Post on 04-Jun-2018
TKK Dissertations 63Espoo 2007
BIOFUEL DRYING AS A CONCEPT TO IMPROVE THE ENERGY EFFICIENCY OF AN INDUSTRIAL CHP PLANTDoctoral Dissertation
Helsinki University of TechnologyDepartment of Mechanical EngineeringLaboratory of Energy Economics and Power Plant Engineering
Henrik Holmberg
TKK Dissertations 63Espoo 2007
Henrik Holmberg
Dissertation for the degree of Doctor of Science in Technology to be presented with due permission of the Department of Mechanical Engineering for public examination and debate in Auditorium K216 at Helsinki University of Technology (Espoo, Finland) on the 13th of April, 2007, at 12 noon.
Helsinki University of TechnologyDepartment of Mechanical EngineeringLaboratory of Energy Economics and Power Plant Engineering
Teknillinen korkeakouluKonetekniikan osastoEnergiatalouden ja voimalaitostekniikan laboratorio
BIOFUEL DRYING AS A CONCEPT TO IMPROVE THE ENERGY EFFICIENCY OF AN INDUSTRIAL CHP PLANTDoctoral Dissertation
Distribution:Helsinki University of TechnologyDepartment of Mechanical EngineeringLaboratory of Energy Economics and Power Plant EngineeringP.O. Box 4100FI - 02015 TKKFINLANDURL: http://www.tkk.fi/Yksikot/Energiatalous/Tel. +358-9-451 3621Fax +358-9-451 3674E-mail: helena.eklund@tkk.fi
© 2007 Henrik Holmberg
ISBN 978-951-22-8648-5ISBN 978-951-22-8649-2 (PDF)ISSN 1795-2239ISSN 1795-4584 (PDF) URL: http://lib.tkk.fi/Diss/2007/isbn9789512286492/
TKK-DISS-2270
Otamedia OyEspoo 2007
AB
HELSINKI UNIVERSITY OF TECHNOLOGY P. O. BOX 1000, FI-02015 TKK http://www.tkk.fi
ABSTRACT OF DOCTORAL DISSERTATION
Author Henrik Holmberg
Name of the dissertation Biofuel drying as a concept to improve the energy efficiency of an industrial CHP plant.
Date of manuscript August 9, 2006 Date of the dissertation April 13, 2007
Monograph X Article dissertation (summary + original articles)
Department Mechanical Engineering Laboratory Energy Economics and Power Plant Engineering Field of research Industrial Energy Engineering Opponent(s) Prof. Thore Berntsson, Dr. Pat McKeough Supervisor Prof. Pekka Ahtila (Instructor) Prof. Markku Lampinen
Abstract This thesis presents the results of a research project in which the enhancement of the CHP production by means of biofuel drying at an integrated pulp and paper mill has been explored. Instead of high temperature flue gas drying the use of low-temperature secondary heat as drying energy has been studied. The drying system considered in this thesis is classified as follows: Drying medium: Air Heat supply into the material: Convection Transport mechanism of the material inside the dryer: Conveyor Method to improve energy efficiency in the drying system: Multi-stage drying. In addition to secondary heat, the option of using steam as a drying energy has been included in the calculation models. An optimization model for analyzing the integration of a multi-stage drying system into the CHP process has been developed. A simulation model based on the optimization model has been created to analyze the operation of the dryer under variable drying conditions. The simulation model can also be applied to the model-based control of the final fuel moisture content if there is a need to control it. Both models have been applied to a case study. Drying as a physical phenomenon has been studied experimentally in a small fixed bed dryer, and guidelines for dimensioning a continuous conveyor dryer in the case of multi-stage drying are presented. The energy efficiency of drying has also been analyzed using two evaluation methods: specific heat consumption and the irreversibility rate. According to case study results, an optimally designed dryer uses only secondary heat, not steam, as a heat source, and earnings stem from decreased marginal fuel consumption, not increased power generation. Simulation calculations show that there are no economic grounds for control of the final fuel moisture content by adjusting heat inputs into the dryer. To control the final fuel moisture content the use of homogenous, and not just dry, fuel must have some positive influences on the operation of the boiler and/or power plant. Experimental tests show that critical moisture content is high for woody-based fuels and diffusion-based drying models must be used to determine drying times theoretically in a fixed bed. If the energy used in drying can be converted to mechanical work, the irreversibility rate is a more comprehensive method of comparing energy efficiency between different drying processes than specific heat consumption.
Keywords Biofuel drying, CHP production, Secondary heat
ISBN (printed) 978-951-22-8648-5 ISSN (printed) 1795-2239
ISBN (pdf) 978-951-22-8649-2 ISSN (pdf) 1795- 4584
ISBN (others) Number of pages 63 p. + app. 62 p.
Publisher Helsinki University of Technology, Dept. of Mech. Eng., Lab. of Energy economics and Power Plant Eng.
Print distribution Laboratory of Energy Economics and Power Plant Engineering
X The dissertation can be read at http://lib.tkk.fi/Diss/2007/isbn9789512286492/
AB
TEKNILLINEN KORKEAKOULU PL 1000, 02015 TKK http://www.tkk.fi
VÄITÖSKIRJAN TIIVISTELMÄ
Tekijä Henrik Holmberg
Väitöskirjan nimi Biopolttoaineen kuivauksen mahdollisuudet parantaa energiatehokkuutta metsäteollisuuden CHP-tuotannossa
Käsikirjoituksen jättämispäivämäärä 9.8.2006 Väitöstilaisuuden ajankohta 13.4.2007
Monografia X Yhdistelmäväitöskirja (yhteenveto + erillisartikkelit)
Osasto Konetekniikka Laboratorio Energiatalous ja voimalaitostekniikka Tutkimusala Teollisuuden energiatekniikka Vastaväittäjä(t) Prof. Thore Berntsson, Tohtori Pat McKeough Työn valvoja Prof. Pekka Ahtila (Työn ohjaaja) Prof. Markku Lampinen
Tiivistelmä Väitöstyössä esitetään tuloksia ja johtopäätöksiä tutkimusprojektista, jossa on tutkittu yhdistetyn sähkön ja lämmöntuotannon (CHP-tuotanto) energiatehokkuuden parantamista biopolttoaineen kuivauksen avulla integroidussa sellu- ja paperitehtaassa. Perinteisen savukaasukuivauksen sijasta tutkimuksessa on analysoitu matalalämpöisen sekundäärilämmön soveltuvuutta biopolttoaineen kuivaukseen. Laskentamallit on muodostettu kuivaussysteemille, jossa kuivauskaasuna toimii ilma, lämmönsiirto kuivauskaasusta materiaaliin tapahtuu konvektiolla, materiaalin siirto kuivurissa toteutetaan jatkuvatoimisella kuljettimella ja kuivurin energiatehokkuutta voidaan tarvittaessa parantaa ilman vaiheistuksen avulla (ns. monivaihekuivaus). Sekundäärilämmön lisäksi laskentamalleissa huomioidaan mahdollisuus käyttää vastapaine- ja/tai väliottohöyryä kuivausilman lämmityksessä.
Työssä on muodostettu optimointimalli monivaihekuivurin liittämiseksi CHP-prosessiin sekä simulointimalli kuivurin toiminnan analysoimiseksi muuttuvissa kuivausolosuhteissa. Simulointimallia voidaan tarvittaessa käyttää polttoaineen loppukosteuden mallipohjaisessa säätämisessä. Sekä optimointi- että simulointimallia on sovellettu esimerkkitapauksissa, joiden lähtöarvot ovat peräisin eräältä suomalaiselta sellu- ja paperitehtaalta. Kuivumista fysikaalisena ilmiönä on tutkittu kokeellisesti pienessä laboratoriokokoluokan kiintopetireaktorissa, ja koetuloksiin perustuen esitetään laskentaperiaatteet kuivurin mitoittamiseksi monivaihekuivauksessa. Työssä on lisäksi analysoitu kuivurin energiatehokkuutta sekä ominaisenergiankulutuksen että exergia-menetelmän perusteella.
Tulosten perusteella optimaalisesti mitoitettu kuivuri käyttää useimmissa tapauksissa lämmönlähteenä ainoastaan sekundäärilämpöä ja kuivauksen tuotot muodostuvat kokonaisuudessaan marginaalipolttoineen kulutuksen vähenemisestä, eivätkä lisääntyneestä sähköntuotannosta. Simulointilaskelmissa tehtyjen oletusten perusteella polttoaineen loppukosteutta ei ole perusteltua säätää vaan sen kannattaa antaa mieluummin vaihdella. Loppukosteuden säätö on perusteltua, jos tasalaatuisen, ei siis pelkästään kuivatun, polttoaineen käytöllä on joitain positiivisia vaikutuksia kattilan ja/tai voimalaitoksen toimintaan. Näitä mahdollisia vaikutuksia ei ole työssä analysoitu. Kokeelliset mittaukset osoittavat, että kriittinen kosteuspitoisuus on korkea puuperäisille materiaaleille ja kuivumisaikojen teoreettisessa laskennassa on käytettävä diffuusioteoriaan perustuvia kuivumismalleja. Jos kuivaukseen käytettävää energiaa voidaan muuttaa mekaaniseksi työksi alueella, jossa kuivuri sijaitsee, pitäisi erilaisten kuivureiden energiatehokkuutta verrata mieluummin exergia-menetelmään perustuen kuin ominaisenergiankulutusten perusteella
Asiasanat Biopolttoaineen kuivaus, CHP-tuotanto, sekundäärilämpö
ISBN (painettu) 978-951-22-8648-5 ISSN (painettu) 1795-2239
ISBN (pdf) 978-951-22-8649-2 ISSN (pdf) 1795- 4584
ISBN (muut) Sivumäärä 63 s. + julkaisut 62 s.
Julkaisija Teknillinen korkeakoulu, Konetekniikan osasto, Energiatalouden ja voimalaitostekniikan laboratorio
Painetun väitöskirjan jakelu Energiatalouden ja voimalaitostekniikan laboratorio
X Luettavissa verkossa osoitteessa http://lib.tkk.fi/Diss/2007/isbn9789512286492/
Preface This work has been done at the Laboratory of Energy Economic and Power Plant Engineering (EVO), Helsinki University of Technology. The study has been funded by Pohjolan Voima Oy, Stora Enso Oyj, Vapo Oy and National Technology Agency of Finland (Tekes). All financers are gratefully acknowledged for making this work possible. I would like to express my gratitude to my supervising professor Pekka Ahtila for his guidance, comments and several good discussions during this work. In some cases, these discussions have had nothing to do with drying or energy efficiency but they have also been important and pleasant. Professor Markku Lampinen has been my instructor in this work and I want to thank him for useful comments and also for a good basic teaching that his has given on his courses. The working atmosphere at the laboratory of EVO has been pleasant and encouraging during these years. Thanks to all current and ex-colleagues for that. Special thanks to Ilkka H. and Mari for several valuable comments to my scientific output and for helping me in different kinds of practical problems. Finally, I want to thank my family members and friends. Even if most of you have had nothing to with the thesis itself it would have been much harder to complete this long process without you. Helsinki, February 2007 Henrik Holmberg
2
Contents PREFACE......................................................................................................................................................... 1 CONTENTS...................................................................................................................................................... 2 LIST OF PAPERS............................................................................................................................................ 3 NOMENCLATURES....................................................................................................................................... 4 1 INTRODUCTION .................................................................................................................................. 6
1.1 BIOFUELS IN THE FOREST INDUSTRY ................................................................................................ 6 1.2 CLASSIFICATION OF BIOFUEL DRYERS .............................................................................................. 9 1.3 BACKGROUND OF BIOFUEL DRYING IN THE FOREST INDUSTRY ....................................................... 11 1.4 OBJECTIVE AND SCOPE OF THE THESIS ........................................................................................... 12 1.5 LITERATURE REVIEW...................................................................................................................... 14
2 DRYING AS A PHYSICAL PHENOMENON .................................................................................. 16 2.1 THEORY ......................................................................................................................................... 16 2.2 EXPERIMENTAL TESTS.................................................................................................................... 17 2.3 CONCLUSIONS ON EXPERIMENTAL TESTS ....................................................................................... 21
3 DESIGN OF THE DRYING SYSTEM............................................................................................... 23 3.1 MULTI-STAGE DRYING SYSTEM ...................................................................................................... 23 3.2 DETERMINATION OF DRYER DIMENSIONS ....................................................................................... 25 3.3 CONCLUSIONS ON THE DRYER DESIGN............................................................................................ 30
4 INTEGRATION AND OPERATION OF THE BIOFUEL DRYER IN AN INDUSTRIAL CHP PLANT ............................................................................................................................................................ 31
4.1 INDUSTRIAL CHP PLANT................................................................................................................ 31 4.2 INTEGRATION OF THE DRYER INTO THE CHP PROCESS ................................................................... 32
4.2.1 Model description ..................................................................................................................... 32 4.2.2 Results and discussion .............................................................................................................. 34
4.3 OPERATION OF THE DRYER IN AN INDUSTRIAL CHP PLANT............................................................ 40 4.3.1 Model description ..................................................................................................................... 40 4.3.2 Method for on-line determination of initial fuel moisture content............................................ 42 4.3.3 Results and discussion .............................................................................................................. 44
4.4 CONCLUSIONS ON INTEGRATION AND OPERATION OF THE DRYER .................................................. 47 5 EVALUATION OF ENERGY EFFICIENCY IN DRYING............................................................. 48
5.1 SINGLE- STAGE DRYING WITH PARTIAL RECYCLING OF DRYING AIR ............................................... 48 5.2 EVALUATION METHODS ................................................................................................................. 49 5.3 RESULTS AND DISCUSSION ............................................................................................................. 50 5.4 CONCLUSIONS ON EVALUATION OF ENERGY EFFICIENCY ............................................................... 53
6 CONCLUDING REMARKS ............................................................................................................... 54 7 RECOMMENDATIONS FOR THE FUTURE RESEARCH AND DEVELOPMENT OF THE DRYER............................................................................................................................................................ 55 REFERENCES ............................................................................................................................................... 56
3
List of papers The thesis is based on the following six papers I Holmberg H, Ahtila P. Drying Phenomena in a Fixed Bed Under Biofuel Multi
Stage Drying. In: Oliveira A, Afonso C, Riffat S, editors. Proceedings of the 1st
International Conference of Sustainable Energy Technologies, Porto, Portugal; June 12-14, 2002. p. EES1 6-11.
II Holmberg H, Ahtila P. Comparison of drying costs in biofuel drying between multi-stage and single-stage drying. Biomass&Bioenergy 2004; 26: 515-530.
III Holmberg H, Ahtila P. Optimization of the bark drying process in combined heat
and power plant process of pulp and paper mill. In: Odilio AF, Eikevik TM, Strommen I, editors. Proceedings of the 3rd Nordic Drying Conference, Karlstad, Sweden; June 15-17, 2005.
IV Holmberg H, Ahtila P. Adjusting of temperature levels in multi stage drying system
by means of outlet air measurements. In :Oliveira A, Afonso C, Riffat S, editors. Proceedings of the 1st International Conference of Sustainable Energy Technologies, Porto, Portugal; 12-14 June, 2002. p. EES2 1-5.
V Holmberg H, Ahtila P. Simulation model for the model-based control of a biofuel
dryer at an industrial combined heat and power plant. Drying Technology 2006; 24:1547-1557.
VI Holmberg H, Ahtila P. Evaluation of energy efficiency in biofuel drying by means
of energy and exergy analyses. Applied Thermal Engineering 2005; 25: 3115-3128. All papers are independent research carried out and written by the author. Professor Ahtila has contributed to the theoretical content of the papers. He has also commented all papers.
4
Nomenclatures A Cross-sectional area of the dryer [m2] Ae Evaporation surface [Ae] b Energy price [€/MWh] C Cost [€] cp Specific heat capacity [J/kgK] DAB Diffusion coefficient of vapour-air mixture [m2/s] Deff Effective diameter [m] E Exergy [W] q Specific heat consumption [J/kg] L Length of conveyor [m] Le Lewis number [-] I Irreversibility rate [W] lv Vaporisation heat of water [J/kg]
dmM& Dry mass flow of fuel [kg/s] Mv Molecular mass of water [kg/mol]
dam& Dry mass flow of air [kg/s]
em& Evaporation rate [kg/s] m Mass [kg] NI Net income [€] NPV Net present value [€] Nu Nusselt number (αDeff/λ) [-] po Total pressure [Pa] pv Partial pressure of vapour [Pa] pv’ Saturated vapour pressure [Pa] Pr Prandtl number (ηcp/λ) [-] Re Reynolds number (vDeff/ν) [-] s Entropy [J/kgK] To Temperature of surrounding [K] T,t Temperature [K,oC] u Fuel moisture content [kg/kgdm] V Volume [m3] W Work [W] v Velocity [m/s] x Air moisture content [kg/kgda] Z Bed thickness [m] Greeks α Heat transfer coefficient [W/m2K] ε Volume fraction of air in the fuel bed [-] Φ Heat flux, heat consumption [W] μ Dynamic viscosity [kg/ms]
5
λ Heat conductivity [W/mK] ρ Density [kg/m3] ν Kinematic viscosity [m2/s] τu Drying time [s] τop Operating hours [h] Δτ Length of time interval [s] Subscripts a Air bs Backpressure steam c Conveyor d Dryer da Dry air db Dry bulb dm Dry mass e Electricity es Extraction steam f Fuel in Inlet, initial mf Marginal fuel out Outlet, final s Surface sh Secondary heat v Vapour wb Wet bulb
6
1 Introduction
1.1 Biofuels in the forest industry Classification and properties of biofuels Material of biological origin excluding material embedded in geological formations and transformed to fossil is called biomass [1]. Biomass can be further divided into woody and herbaceous biomasses. Fuels produced directly or indirectly from these biomasses are called biofuels [1]. All biofuels consumed by the Finnish forest industry are obtained from woody biomasses and can be divided into three main groups: black liquor, sludge, and solid biofuels. Figure 1 illustrates sources from which biofuels are obtained in the forest industry. Black liquor is the most important biofuel in Finland [2]. Sludge streams are usually small compared with other biofuel streams at pulp and paper mills. The third group, solid biofuels, is composed of bark, forest residues, sawdust, cutter shavings, and recycled wood-based waste (e.g. construction wood, crushed or chipped used wood, used paper and board) [3,4]. In 2004, the Finnish forest industry consumed approximately 60,000 TJ solid biofuels as mill fuels [2,5,6]. This thesis focuses on the drying of solid biofuels, and hereinafter the term biofuel always means this group. Figure 2 shows examples of the most common biofuels used by the forest industry in Finland.
Grinding
Refining Pulp
processingPaper making
Sawmill
Integrateddrying and
energy production
Cooking
Drying
Deinking
Waste water treatment
Final felling
Thinning
Paper
PulpSawn goods
Wood handling
Chipping
Merchantable wood
Pulp
Forest residue
Energy wood
Black liquor
Bark
Sawdust, cutter shavings etc.
Recycling paper
To pulp
Sludge
Sludge Waste water
Recycled wood-based waste
Products Waste water Biofuels
Wood-based panel millWood-based panels
Figure 1. Sources of biofuels in the forest industry, reproduced from [4].
7
Bark Forest residue
Crushed constructionwood
Sawdust
20mm 20mm
20mm 20mm
Figure 2. Examples of the most common biofuels used by the forest industry in Finland. The average moisture content of bark is typically between 55-60% w.b. depending on the bark type [7]. In addition to the high average moisture content, the weather, season, and storage time may cause drastic deviations in the bark moisture. The moisture content of sawdust may vary from air-dry to even 70% w.b. [8]. On the other hand, cutter shavings usually have a low moisture content of 5 and 15 % w.b. [8]. The moisture content of the used wood-based products depends on its source. Usually, the material is relatively dry. Forest residue consists of wood material that is left in forest after logging or thinning. Forest residue is crushed before combustion. The moisture content depends on the logistic chain. If the forest residue is delivered directly to the mill, the fuel has almost the same moisture content as the fresh wood. If the residue is allowed to dry after harvesting, the moisture content may decrease even to 20-30 % w.b. during the summer months [8].
8
Benefits of drying
ll boilers used for the combustion of biofuels in Finland are fluidized bed boilers: either
• Improvement of the effective heating value ng from the variation in fuel moisture
• boiler dimensions solid and gaseous compounds from the boiler
he effective heating value (also known as the lower heating value) is directly proportional
he basic functions of the combustion control and burner management systems are to
oiler dimensions may be reduced if the boiler is designed for dry fuel. However, this
enerally, increasing moisture content means a more incomplete combustion. For example,
Athe bubbling fluidized bed boiler (BFB boiler) or the circulating fluidized bed boiler (CFB boiler). Fluidized bed boiler technology enables the combustion of moist fuels, and at present biofuels are not yet dried before combustion in the forest industry. Compared with moist biofuel the use of dry fuel would offer the following benefits:
• Homogenization of the heating value resulticontent Smaller
• Lower amounts of unburned
Tto the fuel moisture content. As a result of drying, energy input into the boiler may be increased without increasing the fuel input, or the fuel input into the boiler may be decreased to get the same energy input as in the case of moist fuel. Tmaintain constant steam flow or pressure under varying loads through proper input of fuel and to maintain safe and efficient operation throughout the boiler’s load range [9]. Compared with several other fuels (e.g. oil, natural gas and coil) the heating value of biofuel varies constantly as a result of varying moisture content. It is possible to control large changes in fuel quality in fluidized bed combustion, but such boilers set high requirements for the process control [10]. At the moment, the trend seems to be towards more advanced process control systems in the boiler [11]. However, drying can be seen as a conceivable method of decreasing the need to control the boiler caused by the varying fuel quality. The need to control the final fuel moisture content in the dryer is discussed in Chapter 4.3. Brequires that the operation of the dryer is sufficiently robust. If moist fuel cannot be dried all the time, a boiler designed for dry fuel may become a bottleneck in steam generation, or fossil support fuels must be used. Gincreasing fuel moisture content means higher emissions of hydrocarbons due to an incomplete combustion [12]. However, it is difficult to obtain any unequivocal correlation between fuel moisture content and emissions, since the type of equipment and the mode of operation also affect the results [12].
9
1.2 Classification of biofuel dryers Biofuel drying systems may be classified using three principal factors:
1. Drying medium Flue gas Air Steam
2. Heat supply into the material Convection (direct dryers) Conduction (indirect dryers) Combination of direct and indirect dryers
3. Transport mechanism of the material inside the dryer Rotary/drum dryers Conveyor dryers (also known as fixed bed dryers) Paddle/screw dryers Fluidized bed dryers Cascade dryers Pneumatic dryers
Further sub-classification to continuous or batchwise dryers is possible but usually unnecessary. Table 1 shows a short summary of the most common direct dryers used in biofuel drying based on references [12-21].
10
Tabl
e 1.
Sum
mar
y of
the
mos
t com
mon
dir
ect d
ryer
s use
d in
bio
fuel
dry
ing
[12-
21]
Dry
er ty
pe
Typi
cal
dryi
ng
med
ium
Rep
orte
d dr
ying
te
mpe
ratu
res
Rep
orte
d ev
apor
atio
n ra
tes
Rep
orte
d he
at a
nd
elec
trici
ty c
onsu
mpt
ions
R
epor
ted
good
side
s [1
4,17
,18,
19]
Rep
orte
d pr
oble
ms/
bad
side
s [14
,17,
18,1
9]
Cur
rent
supp
liers
Dru
m d
ryer
s Fl
ue g
as
200-
600o C
[13]
3.
6-20
t H2O
/h[1
4]
Hea
t 3.
5-4.
2 M
J/kg
H2O
[16]
Elec
trici
ty
10-5
0 kW
h/t dm
[16]
- Sui
tabl
e fo
r fue
ls w
ith
hete
roge
neou
s par
ticle
si
ze
- Rob
ust
- Low
mai
nten
ance
cos
ts
- Dus
t and
smel
l pro
blem
s - T
oo c
oars
e ba
rk h
as
caus
ed b
lock
age
- Fire
risk
afte
r the
dry
er
and
in sh
utdo
wn
GEA
, To
rkap
para
ter,
Dry
Co
Con
veyo
r dr
yers
(als
o kn
own
as
fixed
bed
dr
yers
)
Air
30-1
50o C
[14,
18]
0.5-
40t H
2O/h
[14,
18]
- S
uita
ble
for f
uels
with
he
tero
gene
ous p
artic
le
size
- S
uita
ble
for l
ow
tem
pera
ture
dry
ing
with
lo
ng re
side
nce
times
- R
obus
t - G
ood
cont
rolla
bilit
y
- Lar
ge d
ryer
dim
ensi
ons
- Fire
risk
insid
e th
e dr
yer
Swis
s Com
bi,
Bru
ks K
löck
ner,
Mab
arex
, A
ndrit
z Fi
ber
Dry
ing
Cas
cade
dry
ers
Flue
gas
160-
280o C
[13]
0.
8-7t
H2O
/h[1
4]
Hea
t 5.
8 M
J/kg
H2O
(sha
re o
f flu
e ga
s los
s 2.5
MJ/
kgH
2O
)[16]
Elec
trici
ty
10-1
5 kW
h/t dm
[16]
- Sui
tabl
e fo
r fue
ls w
ith
hete
roge
neou
s par
ticle
si
ze
- Rea
sona
ble
drye
r di
men
sion
s - R
obus
t in
mos
t cas
es
- Cor
rosi
on a
nd e
rosi
on
- Fire
risk
afte
r the
dry
er
and
in sh
utdo
wn
- Lon
g ba
rk st
ripes
hav
e en
tang
led
arou
nd m
ovin
g pa
rts
Hav
e be
en
supp
lied
by B
acho
In
dust
ri. C
urre
nt
situ
atio
n un
clea
r.
Pneu
mat
ic
drye
rs (e
.g
flash
and
mill
dr
yers
)
Flue
gas
St
eam
- F
or fl
ue g
as
drye
rs 1
50-
700o C
[13,
15]
- For
stea
m d
ryer
s te
mpe
ratu
re
depe
nds o
n th
e pr
essu
re a
nd
degr
ee o
f sup
er
heat
ing
us
ually
t dr
y > 1
50o C[1
4,19
]
Flue
gas
dry
ers
10-2
6 t H
2O/h
[14,
19]
Stea
m d
ryer
s 6-
30t H
2O/h
[14,
19]
Hea
t (flu
e ga
s)
3.7
MJ/
kgH
2O[1
6]
Elec
trici
ty (f
lue
gas)
60
-120
kW
h/t dm
(in
mill
dr
yers
shar
e of
grin
ding
25
-40k
Wh/
t dm )[1
4,16
]
Hea
t (st
eam
) 40
0-10
00 k
J/kg
H2O
, whe
n he
at g
ener
ated
in th
e dr
yer
is re
cove
red[1
2,14
,20]
- Sm
all d
ryer
dim
ensi
ons
- Flu
e ga
s dry
ers h
ave
been
robu
st
- Hea
t con
sum
ptio
n is
sm
all f
or st
eam
dry
ers,
if he
at g
ener
ated
in th
e dr
yer i
s rec
over
ed
- Not
as s
uita
ble
for l
arge
pa
rticl
es a
s oth
er d
ryer
ty
pes
- Cor
rosi
on a
nd e
spec
ially
er
osio
n pr
oble
ms =
> hi
gh
mai
nten
ance
cos
ts
- Fire
risk
afte
r the
dry
er
and
in sh
utdo
wn
- Lea
kage
s and
fuel
in- a
nd
outp
ut h
ave
caus
ed
prob
lem
s in
stea
m d
ryer
s
GEA
D
ryC
o Ei
nco
Add
ition
al n
otes
: - C
asca
de d
ryer
can
be
rega
rded
as a
kin
d of
app
licat
ion
of tr
aditi
onal
flui
dize
d be
d dr
yer a
nd p
neum
atic
dry
er, a
nd h
ave
been
qui
te w
idel
y us
ed fo
r bio
fuel
dry
ing,
esp
ecia
lly in
Sw
eden
. - F
luid
ised
bed
dry
ers a
re n
ot fa
vour
ed in
bio
fuel
dry
ing,
bec
ause
mos
t bio
fuel
s ful
fil u
nsat
isfa
ctor
ily c
riter
ia (s
ee R
ef. [
21])
set o
n m
ater
ial s
uita
ble
for f
luid
ised
bed
dry
ing.
- P
addl
e/sc
rew
dry
ers r
epre
sent
usu
ally
indi
rect
dry
ers,
and
are
quite
rare
ly u
sed
for d
ryin
g of
solid
bio
fuel
s. H
owev
er, s
ome
appl
icat
ions
of s
crew
/pad
dle
drye
rs u
sed
for s
awdu
st d
ryin
g ar
e m
entio
ned
in [1
4].
11
1.3 Background of biofuel drying in the forest industry Even though biofuel drying is not common before combustion at the moment, commercial biofuel dryers were used at pulp and paper mills in the 1970s and 1980s. The dominant combustion technique for biofuels at that time was grate firing. This type of boiler can handle fuels with varying moisture content, but ideally a moisture content of 30-40% w.b. should be used [22]. The main reason for dryer investments in the 1970s and the 80s was probably the high oil price resulting from two oil crises. In some cases, the moist fuel also decreased the boiler capacity so much that it became reasonable to install a dryer in combination with the boiler. In the 1970s and 1980s, all industrial dryers in the Finnish forest industry were direct flue gas dryers [13]. Flue gases were either taken directly from the boiler or generated in a separate flue gas burner [13]. According to references [13-15] the most common dryer types were drum dryer, cascade dryer and pneumatic dryer. Since the 1970s fluidized bed boilers have replaced grate firing as a combustion technique [7]. Compared with grate firing, the fluidized bed boiler is a more suitable combustion technique for moist biofuels. The moisture content must be higher than approximately 62-65% w.b. before stable combustion becomes difficult to maintain and it becomes necessary to support the firing with some fossil fuel [7]. However, the use of moist fuel decreases the energy efficiency of the power plant, and one may ask why the existing flue gas dryers were not integrated with fluidized bed boilers. At least, the following reasons may be listed:
• Bad operating experiences • Environmental considerations • Economic factors
Biofuel is a burnable material with a heterogeneous particle size. It is also typical that the biofuel flow contains stones, sand and other inappropriate things. It is obvious that the properties of biofuel set tough requirements for the operation of the dryer. For some dryer types (e.g. pneumatic dryers) the maintenance cost may also be high. It is presumable that the operational experiences of flue gas dryers have not always been satisfactory and bad experiences have supported the decision to dispense with the dryer. It is still one of the main prerequisites for drying that the dryer is a robust technique. The advantages and drawbacks of different dryers are listed in Table 1. All types of wood material contain volatile organic material that may be emitted together with the water vapour [12]. There are presently legal restrictions on the amounts that may be released. The emissions of biomass drying are greatly affected by the drying temperature, when it exceeds 100oC [23]. Below 100oC emissions are reported to be low [23, 24]. The drying temperatures of flue gas dryers are clearly higher than 100oC (see Table 1). Exhaust gases or unclean condensates must be treated after the dryer if they contain high concentration of emissions. Treatment increases drying costs. The need for the treatment of the exhaust gas depends on current regulations. After the oil crisis, energy prices fell and dryer investments were no longer so profitable. Due to the absence of economic drivers, attention was not paid to the energy efficiency of
12
CHP plants. It was more important to develop robust combustion technologies. However, economic drivers (e.g. emissions trading, increasing fuel prices) are more favourable for drying at the moment.
1.4 Objective and scope of the thesis In 1998 a research project called “Enhancement of combined heat and power production at integrated pulp and paper mill” began at the Helsinki University of Technology. The main objective of the project was to study how to enhance combined heat and power production (CHP production) by means of drying at integrated pulp and paper mills. Instead of high-temperature flue gas drying, the idea was to use low-temperature secondary heat as drying energy. The following research questions were set in the project:
1. Utilization of secondary heat in drying 2. Methods to improve the energy efficiency of the dryer 3. Methods to decrease the drying gas demand without increasing the drying
temperature 4. Evaluation of quality and quantity of emissions released from biofuels with low-
temperature drying 5. Integration of a biofuel dryer into the CHP process in an optimal way 6. Operation of biofuel dryer under variable drying conditions (integrated into the
CHP plant) and the need of control final fuel moisture content In general, secondary heat means the heat which is expelled from a process in own streams without the heat of the product [25]. Secondary heat from processes goes into the secondary heat system, which is a water-based heat supply system [25]. Temperatures of secondary heat flows are usually lower than 100oC, which means that air is the only realistic drying gas. Heat from secondary heat flows is transferred to air in indirect heat exchangers. In addition, backpressure and/or extraction steam may be used for the heating of air to get higher drying temperatures, if it is reasonable. A conveyor/fixed bed dryer is particularly recommended for low-temperature drying (see Table 1) [14,18]. Several flow sheets for decreasing the heat consumption in air drying may be devised [26]. In the above-mentioned research project, a multi-stage drying system (MSD) was selected as the primary method of improving the energy efficiency in drying and reducing the drying air demand. Dr Jukka-Pekka Spets worked as a research scientist in the above-mentioned project, and his doctoral thesis “Enhancement of The Use of Wood Fuels in Heat and Power Production in Integrated Pulp and Paper Mill [27]” was published in 2003. In [27], an in-depth performance analysis of the multi-stage drying system was made from thermodynamic point of view. Qualitative and quantitative analyses of emissions released from biofuels with low temperature drying based on experimental tests have also been presented in [27] .
13
The drying system considered in this thesis is classified as follows: Drying medium: Air Heat supply into the material: Convection Transport mechanism of the material inside the dryer: Conveyor Method to improve energy efficiency: Multi-stage drying. Both secondary heat and steam can be used for heating of drying air. The main objective of this thesis is to analyze research questions 5 and 6. An optimization model for analyzing the integration of a multi-stage drying system into the CHP-process has been developed. A simulation model based on the optimization model has been created to analyze the operation of the dryer under variable drying conditions. The simulation model can also be applied to model-based control of the dryer. Conclusions on the need to control the final fuel moisture content are drawn by comparing simulation results, which are calculated for dryers with and without control. Both models are applied to a case study and results for the integration and operation of the dryer are presented in Chapters 4.2 and 4.3, respectively. In connection with the simulation model, a simple method for approximately determining the initial fuel moisture content is also presented. To be able to analyze research questions 5 and 6, the drying phenomenon of wood-based material must be known, and guidelines for dimensioning the chosen drying system must be determined. To study the drying phenomenon a series of experimental drying testes have been conducted in a laboratory. Results of experimental tests are presented in Chapter 2. Guidelines for dimensioning a continuous conveyor dryer in the case of multi-stage drying are presented in chapter 3. Finally, evaluation methods for energy efficiency are discussed. In Chapter 5, energy efficiency of multi-stage drying is analyzed using two evaluation methods: Specific heat consumption and the irreversibility rate. Theoretical minimums for heat consumptions and irreversibility rates are presented and the main difference between the two evaluation methods is clarified. The energy efficiency of multi-stage drying is also compared to single-stage drying with partial recycling of drying air, which is an alternative flow sheet for improving the energy efficiency in drying. The main results of this thesis are presented in Chapters 2-5, and they are based on six appendix papers: I Holmberg H, Ahtila P. Drying Phenomena in a Fixed Bed Under Biofuel Multi
Stage Drying. In: Oliveira A, Afonso C, Riffat S, editors. Proceedings of the 1st
International Conference of Sustainable Energy Technologies, Porto, Portugal; June 12-14, 2002. p. EES1 6-11.
II Holmberg H, Ahtila P. Comparison of drying costs in biofuel drying between multi-stage and single-stage drying. Biomass&Bioenergy 2004; 26: 515-530.
III Holmberg H, Ahtila P. Optimization of the bark drying process in combined heat and power plant process of pulp and paper mill. In: Odilio AF, Eikevik TM, Strommen I, editors. Proceedings of the 3rd Nordic Drying Conference, Karlstad, Sweden; June 15-17, 2005.
IV Holmberg H, Ahtila P. Adjusting of temperature levels in multi stage drying system by means of outlet air measurements. In :Oliveira A, Afonso C, Riffat S, editors.
14
Proceedings of the 1st International Conference of Sustainable Energy Technologies, Porto, Portugal; 12-14 June, 2002. p. EES2 1-5.
V Holmberg H, Ahtila P. Simulation model for the model-based control of a biofuel dryer at an industrial combined heat and power plant. Drying Technology 2006; 24:1547-1557.
VI Holmberg H, Ahtila P. Evaluation of energy efficiency in biofuel drying by means of energy and exergy analyses. Applied Thermal Engineering 2005; 25: 3115-3128.
1.5 Literature review The aim of this chapter is to give a brief review of the most relevant research projects related to biofuel drying under the topic of this thesis. Because there are several pulp and paper mills in Finland and Sweden, and both countries use a lot of biofuels, the most relevant studies are done either in Finland or in Sweden. Especially, the Swedish foundation Värmeforsk has funded a lot of studies dealing with the forest industry and biofuels. References [28-30] discuss utilization of secondary heat at integrated pulp and paper mills based on five case studies. The papers present secondary heat balances of the mills, proposals for more efficient use of secondary heat, and economic analyses of secondary heat investments. There are substantial amounts of unused secondary heat available in all mills. For example, in the mill analyzed in [29] there are hot waters at the temperature of 80oC and temperature range of 57-70oC available 44kg/s and 171kg/s, respectively. However, all available secondary heat flows cannot be used for all purposes. Utilization is always mill-specific, depending on the temperature of the secondary heat flows, mill layout, process concepts, duration curves etc. Biofuel drying is not mentioned as an option for the utilization of secondary heat in [28-30]. In [18], the technical and economic aspects of utilizing secondary heat for biofuel drying has been studied in cases representing both saw mills and pulp and paper mills. A survey of available secondary heat sources at pulp and paper mills and has been made, and commercial drying technologies suitable for biofuel drying have been discussed. The profitability of drying has also been evaluated for single-stage drying. The study resulted in the following main conclusions: 1) There are large sources of secondary heat available for drying purposes at pulp and paper mills. 2) A continuous fixed bed dryer seems to be the most suitable drying technology for biofuel drying at low drying temperatures. 3) Payback periods for the dryer investment are between 2.5-3.4 years depending on the selected initial values. The results presented in [18] support the results of our study. However, several issues discussed in this thesis (e.g. multi-stage drying, design guidelines for the fixed bed dryer, and integration of the dryer into the CHP process) are beyond the scope of [18]. In [17] and [19], operational experiences of existing biofuel dryers have been studied. In most cases, drying is related to pellet manufacturing processes. Both steam and flue gas/air dryers have been analyzed in [17] and [19]. The studies pose questions about raw material, feeding and discharge systems of dried goods, gas and dust handling, fires and control
15
system. According to the results, flue gas/air dryers are more robust than steam dryers. Corrosion and erosion problems and leakage from the feeding and discharge systems have been the most typical problems of steam dryers. From the viewpoint of this thesis, the most relevant information relating to operation experiences is that an air dryer operating at a drying temperature of 100-120oC has had no environmental problems or fires [17]. The dryer is a continuous fixed bed dryer and is used for bark drying at Rockhammars Bruk in Sweden. In references [31-33], and also in reference [19], the authors discuss control methods for both flue gas and steam dryers used in the pellet manufacturing. In most cases, moisture control of the outgoing material is based on measurement of the exhaust gas temperature. Reference [33] presents a control method based on the connection between the humidity in the flue gas and the moisture in the biofuel. According to references, control methods seem to work quite well. However, the control methods presented for biofuel dryers in [19] and [31-33] are not related to drying in the connection of the CHP plant. They are also based on conventional feedback control, and not on the simulation model as in this book. Reference [34] contains a literature review of methods for determining moisture content in the biofuel flow. Scanning with Radio Frequent Electromagnetic Waves (RF), Near Infrared Spectroscopy (NIR), Microwaves and Nuclear Magnetic Resonance (NMR) are mentioned in [34] as conceivable methods for determining moisture in the biofuel flow. Because of several reasons mentioned in [34] it is hard to reliably compare the different methods. According to [34] NIR technology can be regarded as the most promising method for the determination of moisture in the biofuel flow. In [35], the Technical Research Centre of Finland (VTT) has studied drying of wood-based fuel particles both theoretically and experimentally. The drying models presented are based on the solution of the conservation equations for mass and energy. Theoretical drying models for a single particle and for particles in fixed and moving beds are presented in [36, 37]. In both papers, the models are presented as a one-dimensional case for platy, cylindrical, or spherical particles. Experimental tests are carried out by using a thermo balance reactor for drying studies on a single particle. Experimental results are presented in [35-38]. In addition to [35-38], drying of wood-based fuel particles has been studied both theoretical and experimental in [39,40]. Both papers present a two-dimensional drying model for the transport mechanism of a single particle. In [41], a comprehensive heat and mass transfer model is presented for fixed-bed drying of spherical particles. The drying models presented might have been useful methods to calculate drying times in this study, too. However, drying times for the chosen biofuels have been determined experimentally. The decision to opt for experimental tests rather than theoretical models is explained in Chapter 2.3.
16
2 Drying as a physical phenomenon
2.1 Theory All biofuels are porous hygroscopic material. Moisture inside the porous material may be in the frozen, liquid or vapor state; in addition, there may be some non-condensable gas present [42]. The transfer of moisture inside the material may occur by several mechanisms, such as diffusion in continuous homogenous solids, capillary flow in granular and porous solids, flow due to shrinkage and pressure gradients, and flow caused by sequential of vaporization and condensation [43]. It is common that different transport mechanisms predominate at different times in a drying cycle. The factors governing heat and mass transfer determine the drying rate [43]. Drying is usually divided into three successive periods: 1) the warming-up period, 2) the constant rate period, and 3) the falling rate period. The warming-up period is normally short compared with the constant and falling rate periods. In the constant rate period, the surface contains free moisture and vaporization takes place from the surface. Toward the end of the constant rate period, moisture has to be transported from inside of the solid to the surface by capillary forces [26]. In the constant drying period, drying is externally controlled by heat and mass transfer resistance on the gas side only. When the average moisture content reaches the critical moisture content, the falling rate period begins. In the falling rate period, drying is controlled by diffusion of moisture from the inside to the surface and then mass transfer from the surface to the drying medium [26]. Drying is said to be internally controlled. In the constant drying period, heat and mass transfers are in equilibrium at the particle surface. If heat is supplied by convection from hot air, the drying rate may be calculated as follows:
esv
sae A
)t(l)tt(
m−
=α
& , (1)
where α is the heat transfer coefficient, ta the air temperature, ts the surface temperature of the particle, Ae the evaporation surface and lv(ts) vaporization heat of the water at the surface temperature. In the constant drying period, the surface temperature is the same as the wet bulb temperature. If there are several particles in a fixed-bed, and all particles have reached the local wet bulb temperature, the temperature difference in (1) can be replaced as a logarithmic mean value temperature:
)t(t)t(t
In
)t(t)t(tΔt
s2outa,
s1ain,
s2outa,s1ina,ln
−−
−−−= , (2)
17
where ta,in is the inlet air temperature, ta,out the outlet air temperature, ts1 the surface temperature in the bottom part of the bed, and ts2 the surface temperature in the upper part of the bed. Based on the analogy between heat and mass transfer and the diffusion theory in the boundary layer, the theoretical wet bulb temperature may be derived from Equation (1). The derivation is presented, for example, in [44], and the wet bulb temperature may be calculated from the following equation [44]:
)(ln)( ,
1
wbvo
vowbv
no
p
vwba tpp
pptlLe
RTp
cM
tt−−
=− −
ρ , (3a)
where ρcp = ρdacpda + ρvcpv , (3b)
Le = ABp D
cλρ
, (3c)
where ρ is the density, cp the specific heat capacity, Le the Lewis number, DAB the diffusion coefficient of the vapour-air mixture and λ the heat conductivity of the vapour-air mixture. The vapour pressure pv’ as a function of temperature may be expressed using an empirical correlation. Reference [44] gives the following correlation:
230)64.99(78.11
5' 10 +−
= wb
wb
tt
v ep [Pa] , (4) The vaporization heat of water as a function of the vaporization temperature may be given approximately as follows [43]: lv(twb) = 2501 – 2.34twb [kJ/kg] (5) Substituting (4) and (5) in (3a), the wet bulb temperature is the only unknown parameter, and can be calculated from (3a).
2.2 Experimental tests A series of experimental test was conducted. The tests had the following objectives:
• To determine experimentally the heat transfer coefficient in a fixed bed for wood particles
• To evaluate the applicability of the constant drying model for determining the drying rate in a fixed-bed drying
18
The tests were conducted using the test rig shown in Figure 3. To calculate the heat transfer coefficient from Equations (1) and (2), the following parameters must be measured: Absolute air moisture contents and temperatures before and after the bed, surface temperatures of particles in the bottom and upper part of the bed, and the evaporation surface. All particles used in the tests were platy spruce particles of regular shape, and the evaporation surface of the bed was determined by multiplying the surface of a single particle by the number of particles in the bed. Other necessary parameters were measured using thermocouples and humidity transmitters (see a more detailed description of the experimental arrangement in Appendix paper I). The measured temperatures and moisture contents used in Equations (1) and (2) represented averages over the period during which the entire bed was in the constant drying period. In all cases, the constant drying period was relatively short, lasting only some ten seconds (see Appendix paper I).
Reactor∅ 100mm
Pre-heater
Evaporator
Fuel bed
x
x,t
t
m
By pass
Dry air
Vapour
Moist air
Measurement points:x Humidity t Temperaturem Mass flow
400m
m
Figure 3. Drying test rig The particle sizes used in the tests are expressed as effective diameters. The effective diameter is defined as follows [36]:
e
eff AVD 6
= , (6)
where Ae is the total evaporation surface of the particle, and V the volume of the particle. Heat transfer coefficients were determined for four different particle sizes. Effective diameters and actual dimensions (thickness*length*width) of the used particles were 5mm (2.5*10*10mm), 10mm (20*20*5mm), 15mm (7.5*30*30mm), and 20mm (10*40*40mm). Spruce particles were used because most of the raw wood consumed by the forest industry in Finland is softwood. Two different inlet air temperatures, 70 and 120oC were used. With the inlet temperature of 70oC, the inlet air was dry. With the inlet temperature of 120oC the inlet air moisture content was around 20 + 4g/kgd.a. The air was moisturized with an inlet temperature of 120oC to simulate drying conditions corresponding to the multi-stage drying system (a description of the system is presented in Chapter 3.1)
19
The air velocities used in the tests were 0.3 m/s, 0.45 m/s (only with the inlet temperature of 70oC), 0.6 m/s, 0.9 m/s and 1.2 m/s. All velocities were calculated per free sectional area of the grate. Because the air density falls as the air temperature rises, the mass flows were smaller with the inlet temperature of 120oC. The experimentally determined heat transfer coefficients for inlet temperatures of 70oC and 120oC are shown in Figure 4. The measured heat transfer coefficients are smaller for the inlet temperature of 120oC. For both inlet air temperatures, the air velocity before the bed has been the same but the average velocities over the bed have not been the same. As a result of a more drastic temperature drop over the bed, the average velocity has been smaller for the inlet temperature of 120oC. The lower average velocity is probably the main reason, why heat transfer coefficients differ depending on the inlet air temperature.
0
20
40
60
80
100
120
140
160
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Air velocity at inlet temperature [m/s]
Hea
t tra
nsfe
r coe
ffici
ent [
W/m
2 K] Deff =5mm
Deff =10mm
Deff =15mm
Deff =20mm
Inlet air temperature 70°C
0
20
40
60
80
100
120
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Air velocity at inlet temperature [m/s]
Hea
t tra
nsfe
r coe
ffici
ent [
W/m
2 K]
Deff =5mm
Deff =10mm
Deff =15mmDeff =20mm
Inlet air temperature 120°C
Figure 4. Experimentally determined heat transfer coefficients for effective diameters of 5, 10, 15 and 20mm at inlet temperatures of 70oC and 120oC, Deff = effective diameter (see Equation 6). To evaluate the validity of the measured heat transfer coefficients they have been compared to heat transfer coefficients calculated using two different Nusselt number correlations. The first correlation is the Ranz-Marshall correlation (Nu1), and the second one is the Malling and Thodos correlation (Nu2). The correlations are in the following form [36,45]: Nu1 = 2ε + 0.69(Re/ε)1/2(Pr)1/3 (7a) Nu2 = 0.539(Re)0.563(Pr)1/3/ε1.19 (7b) Comparison to these two Nusselt number correlations has been done for cases where the inlet air temperature is 70oC. Reynolds and Prandtl numbers have been determined at the surface temperature of the particle. The surface temperature used represents the average of the measured surface temperatures in the bottom and upper part of the bed. Depending on the measurement, the surface temperatures used varied between 21.5-27.5oC. For small particles surface temperatures are higher. The properties (e.g. viscosity, conductivity) of moist air are taken from [42]. The volume fraction of air ε is 0.57 in the Nusselt number
20
correlations, and has been determined experimentally by filling a measure of volume with particles. The results of the comparison are shown in Figure 5. Because the number of pages was limited in Appendix paper I, Figure 5 was excluded in this paper.
0
20
40
60
80
100
120
140
0 0.2 0.4 0.6 0.8 1
Air velocity [m/s]
Hea
t tra
nsfe
r coe
ffici
ent [
W/m
2 K]
Nu1
Nu2
Measured value Deff =5mm
0
20
40
60
80
100
120
0 0.2 0.4 0.6 0.8 1 1.2 1.4Air velocity [m/s]
Hea
t tra
nsfe
r coe
ffici
ent [
W/m
2 K]
Nu2
Nu1
Measured value
Deff =10mm
0
10
20
30
40
50
60
70
80
90
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Air velocity [m/s]
Hea
t tra
nsfe
r coe
ffici
ent [
W/m
2 K] Nu2
Measured value
Nu1
Deff =15mm
0
10
20
30
40
50
60
70
80
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Air velocity [m/s]
Hea
t tra
nsfe
r coe
ffici
ent [
W/m
2 K] Nu2
Nu1
Measured value
Deff =20mm
Nu1 = 2ε + 0.69(Re/ε)1/2(Pr)1/3 Nu2 = 0.539(Re)0.563(Pr)1/3/ε1.19
Nu1 = 2ε + 0.69(Re/ε)1/2(Pr)1/3 Nu2 = 0.539(Re)0.563(Pr)1/3/ε1.19
Figure 5. Comparison of measured heat transfer coefficients at an inlet air temperature of 70oC with coefficients calculated using Ranz-Marshall correlation (Nu1) and Malling and Thodos correlation (Nu2), Deff = effective diameter (see Equation 6). Figure 5 shows that experimentally measured heat transfer coefficients correspond to calculated heat transfer coefficients reasonably well. Especially for small particles, the Malling and Thodos correlation is good. On the basis of Figure 5 it is reasonable to assume that the validity of the measured heat transfer coefficients is also good for the inlet temperature of 120oC. To evaluate the applicability of the constant drying model for determining the drying rate, the constant drying model was used to calculate the decrease in the moisture content of the particles. The bed was divided into ten layers, and the evaporation surface was assumed to be equal in each layer. In each layer, the drying conditions were assumed to be constant and the air temperature and moisture content of the upper layer was calculated by applying the energy and mass balance of adiabatic moisturizing. Experimentally determined heat transfer coefficients were used in the model. The calculation model is explained in more detail in Appendix paper I.
21
The reduction in the moisture content calculated by the constant drying model was then compared with thee measured value. The effect of air velocity and particle size on the critical sample moisture content was evaluated. In this connection, the critical sample moisture content refers to average sample moisture content below which the constant drying model is no longer accurate. According to the results shown in Appendix paper I, the critical sample moisture content seems to be slightly lower for small particles and air velocities. However, the influences are so small that they can be ignored. Figure 6a shows for all particle sizes a comparison between actual sample moisture contents and theoretical moisture contents calculated using the constant drying model. Each curve in Figure 6a represents already the average of the measurements where air velocity has been a variable. Figure 6b shows the average of the curves shown in Figure 6a, and this curve gives us an estimation of the critical sample moisture content. Figure 6a and b show that the critical sample moisture content is in the order of 1.0 kg/kgdm, and the constant drying model is no longer accurate, since the sample moisture content has decreased below 0.8-0.9kg/kgdm.
0.0
0.4
0.8
1.2
1.6
2.0
0.0 0.5 1.0 1.5 2.0
Theoretical sample moisture [kg/kgdm]
Act
ual s
ampl
e m
oist
ure
[kg/
kgdm
]
5mm(70°C)5mm(120°C)10mm(70°C)10mm(120°C)15mm(70°C)15mm(120°C)20mm(70°C)20mm(120°C)
0.0
0.4
0.8
1.2
1.6
2.0
0 0.5 1 1.5
Theoretical sample moisture [kg/kgdm]
Act
ual s
ampl
e m
oist
ure
[kg/
kgdm
]
2
Figure 6. a) The effect of particle size and temperature on the accuracy of the constant drying model. b) The average accuracy of the constant drying model.
2.3 Conclusions on experimental tests The following conclusions can be drawn from experimental tests
• The stage during which the entire bed is in the constant drying period is short compared with the total drying time.
• Experimentally determined heat transfer coefficients behave logically as a function of particle size and air velocity.
• The Nusselt number correlation of Malling and Thodos determines the heat transfer coefficient with good accuracy for particle sizes of 5 and 10mm.
• The accuracy of the Ranz-Marshall Nusselt number correlation improves as the particle size increases.
22
• The critical moisture content in a fixed bed is in the order of 1.0kg/kgdm for the spruce particles used in the tests.
• The constant drying model is not accurate enough below the critical moisture, and it is necessary to use diffusion-based drying models to determine the drying time theoretically in a fixed bed, even if the final fuel moisture content is clearly higher than the fiber saturation point.
Even if the accuracy of the constant drying model was better for average moisture contents clearly below 0.8-1.0kg/kgdm, the application of the model to actual biofuels would still require further development. The particle size distribution is extremely wide in an actual biofuel flow (see. Figure 2), and it would be necessary to determine the “correct” particle size from the viewpoint of the model. Especially in the case of bark, it is hard to say what the particle size of the fuel is. The use of diffusion-based drying models would require material properties such as diffusivities and conductivities to be determined in addition to the “correct” particle size. In stead of further model development, it was decided that experimentally determined drying curves would be used to determine the drying times.
23
3 Design of the drying system
3.1 Multi-stage drying system The flow sheet of the multi-stage drying system used in the model development is shown in Figure 7. Figure 7b illustrates the states of the drying air on an enthalpy-humidity chart diagram in multi-stage drying. A drying stage consists of the heating and drying period. The addition of drying stages reduces heat consumption in drying, because the air temperature before heating increases constantly as the number of drying stages increases (see Figure 7b) [46]. Adding the drying stages also decreases the air demand in drying if all the drying stages operate at the same temperature level. The reason is the higher outlet air humidity, which is achieved as a result of several drying stages. To reach the same outlet air humidity in single-stage drying, the drying temperature should be higher than in the case of multi-stage drying (see. Figure 7b) [46]. In Figure 7b, two-stage drying gives the same change of air moisture content as the single-stage drying, but air temperatures are lower in two-stage drying. Although the air demand decreases, the dryer dimensions do not become smaller because the drying air must pass through several drying stages. Heating the drying air successively in three heat exchangers does not reduce the heat consumption but decreases the consumption of steam. An optimal combination of heat sources in each drying stage depends on the initial values and is calculated by the model. Thus, there does not have to be three heat exchangers in each drying stage.
24
igure 7. Multi-stage drying a) flow sheet b) states of the drying air on an enthalpy-umidity chart diagram.
Drying stage 1 Drying stage 2 Drying stage n
Air (tin , xin , mda) Air to combustion air (tout , xout , mda)
Φ1 Secondary heat , typical temperature range 50-90 oC Φ3 Back pressure steam , typical temperature range 130-145 oC Φ3 Extraction steam , typical temperature range 180 - 190 oC
Φ1
Φ2
Φ3
Φ1Φ1
Φ2 Φ2
Φ3Φ3
11
12
13
14
21
22
23
24
31n1
n2
n3
n4
1 2 3 n n+1
a)
11
12
13
14
21
22
23
24
n1
n4
1st heating period , Φ1
2nd heating period , Φ2
3rd heating period, Φ3
Drying period
Single-stage drying Heat sources Φ1 Secondary heat Φ2 Backpressure steam Φ3 Extraction steam
Translations: kuiv. ilm.= dry air torr luft = dry air kg vettä/ kg kuiv.ilm. = kg water / kg dry air kg vatten / kg toff luft = kg water / kg dry air
MSD
NOTE! Due to limited temperature scale of the enthalpy humidity chart, drying temperatures in MSD on are lower than actual ones after the heating with secondary heat, backpressure steam and extraction steam.
b)
Fh
25
3.2 Determination of dryer dimensions
ryer dimensions are determined for a continuous conveyor dryer in which air moves
al moisture content is determined experimentally based on the onclusions presented in Chapter 2. In addition to single-stage drying, a concept for etermining dryer dimensions in multi-stage drying is presented.
:
Dthrough a perforated conveyor and a fixed fuel bed (see. Fig. 8). Functionally, the dryer corresponds to a cross-flow concept. The drying time from the given initial moisture ontent to the desired finc
cd
⎯⎯ outlet air moisture------ inlet air moisture
Characteristic length
Air
moi
stur
e
Fuel bed
Figure 8. A continuous cross-flow dryer and the change of outlet air moisture content during drying An equation for determining the mass flow of dry air in continuous cross-flow drying is derived in Appendix paper II, and the final equation takes the following form
udm
dmdaada τ
ε)ρZ(1Mρv
m−
=&
& , (8)
where va is the air velocity per free sectional area of the dryer, ρda the density of dry air, dmM& the dry mass flow of fuel, Z the bed height, ε the volume fraction of air in the fuel bed,
dm the density of dry fuel, and τu the drying time from the given initial moisture content to
ρe desired final moisture content.
, the drying time can be determined directly or indirectly. In this study, the rying time has been determined indirectly by measuring the change of air moisture content
th Experimentallydover the fuel bed situated in a test rig (see Figure 3). The decrease in fuel moisture content is calculated as follows:
∑=−n
iuiiinout ,)xx τΔττΔ∑==
−=i
n
idm
daain (
Zv
uu11ρ
ρ, (9)
where xout is the outlet air moisture content, xin the inlet air moisture, n the number of time vals, Δτ the length of time interval, and τinter u the total drying time. The length of the time
interval was 5 seconds in the experimental tests. Equation (9) represents a drying curve in a mathematical form and the measured curve is analogous to continuous cross-flow drying.
0 1
2 3
1 warm-up period2 period of constant drying rate3 period of falling drying rate
1
Dry air
Moist air Moist fuel Dry fuel
0 1Characteristic length
26
In Equation (8), the air velocity and bed height are known parameters, and the drying
the air velocity. In fixed bed drying, the air velocity must be ightly lower than the minimum fluidized velocity. The minimum fluidized velocity
depends on the particle size, particle density, and volume fraction of aiexample, Reference [47] presents an equation for the e minimum fluidized velocity. In practice, the correct air velocity per free sectional area must
m
city 0.6m/s, inlet air moisture 3±0.5g/kgda, initial
curves are measured for the chosen air velocity and bed height. One of the main questions is how to choose these parameters? In the case of air velocity the answer is simple. The air velocity must be as high as possible, because the cross-sectional area of the dryer is inversely proportional tosl
r in the bed [46]. For theoretical determination of th
be deter ined experimentally for biofuels. Velocities between c. 0.4-0.6m/s are reported for fixed bed dryers in Ref. [14]. In this study, the air velocity at the inlet temperature of drying air is c. 0.6-0.65m/s in the determination of the drying curves (see Figure 10).
quation (8) shows that the air mass flow is inversely proportional to the bed height. On theEother hand, the drying time becomes longer as the bed height increases. To evaluate the effect of bed height on air mass flow, the behavior of the ratio between the drying time and the bed height (τu/Z) was experimentally studied. Regularly shaped wood particles were dried in a fixed bed reactor (see Figure 3) by changing the bed height and keeping the other drying parameters constant. All the particles were ideal spruce particles of the same size and dimensions (2x20x5mm). Two different air temperatures, 70oC and 120oC, were used. Figure 9 shows the results of the tests.
Figure 9. The ratio between drying time and bed height as a function of bed height for inlet temperatures 70oC and 120oC (air velosample moisture 1.7kg/kgdm).
Inlet air temperature 70°C
40
50
60
0
10
20
30
0 50 100 150 200
Bed height [mm]
τ/Z
[s/m
m]
1.51.31.10.90.70.5
Finalsample moisture
Inlet air temperature 120°C
25
30
35 Finalsample moisture
1.51.3
20mm
]
1.10.90.70.5
0
5
10
15
0 50 100 150 200 250
Bed height [mm]
τ/Z
[s/
27
Despite the final moisture content of the sample, the ratio τu/Z seems to decrease constantly as the bed height increases. However, the derivative of the ratio is clearly smaller when the bed heights are over 80-100 m om and 100-120 mm for inlet temperatures of 70 and 120 C,
spectively. Both bed thicknesses approximately represent bed heights when the outlet air fully saturated at the beginning of drying (see Figure 8 in Appendix paper II). On the asis of these measurements, the bed must be at least so high that the drying air reaches its turation point at the beginning of the drying. Increasing the bed height still decreases the
ryer size but the influence on dimensions is not as significant as for thin beds. Bed heights etween 0.2-0.8m are reported for conveyor dryers in Ref. [14]. To achieve a uniform air istribution over the fuel bed it may be necessary to use higher bed heights close to ported ones in actual dryers.
igure 10 shows experimentally determined drying curves when bark (see Figure 2a) has
reisbsadbdre Fbeen dried in a fixed bed dryer. Figure 10a shows drying curves as a function of dry bulb temperature, and Figure 10b shows the same curves as a function of the difference between dry and wet bulb temperatures. Equations (3)-(5) show that the temperature difference depends on both dry bulb temperature and inlet air moisture content. To take into account the effect of air moisture content on drying time, the difference between dry and wet bulb temperatures is used in calculation models instead of merely the dry bulb temperature. Drying time τu for a certain change of fuel moisture content is expressed using a correlation based on experimental determined drying curves. The correlation is defined as follows:
btt
tt a
wbdb
wbdb
u
u +−−
= ])(
ln[maxmax,τ
τ , (10)
where tdb is the dry bulb temperature, twb the wet bulb temperature and τu,max the drying time for the smallest temperature difference (tdb-twb) when fuel is dried from a given initial moisture to a desired final moisture. The term (tdb-twb)max is the greatest temperature difference used in the experiments. Coefficients a and b are calculated from experimental drying curves based on the minimization of the partial least square, which can be easily done by the computer.
28
Figure
mperature b) as a function of the difference between dry and wet bulb temperatures. 10. Experimentally determined drying curves a) as a function of dry bulb
te
Bed height 200mm, air velocity per free-sectional area 0.65m/s
0
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
0 1000 2000 3000 4000 5000
Drying time [s]
Bar
k m
oist
ure
[kg/
kgdm
]
50 °C70 °C90 °C110 °C130 °C150 °C
150°C 130°C 110°C 90°C 70°C 50°C0.0
tdb
a)
b) Bed height 200mm, air velocity per free-sectional area 0.65m/s
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
1.60
0 1000 2000 3000 4000 5000
Drying time [s]
Bar
k m
oist
ure
[kg/
kgdm
]
32 °Ctdb-twb
46 °C61 °C78 °C95 °C110 °C
110°C 95°C 78°C 61°C 46°C 32°C
29
Guidelines for determining the dryer size in the case of single-stage drying are now resented. In multi-stage drying, the final fuel moisture content uout is achieved after the st drying stage, and the fuel moisture content after the previous stages is somewhere etween u -u To use the Equation (8) in multi-stage-drying, the fuel flow between
ture alance over the drying chamber is:
(11)
ubstituting Equation (8) in (11) the air moisture content after the drying chamber can be
plab in out. dm
each drying stage is shared as shown in Figure 11. The change in fuel moisture content for the shared flow in each drying stage is u
M&
in-uout, but the fuel flow is smaller. The moisb
)uu(M)xx(m outindminoutda −=− &&
Scalculated. The inlet air moisture content before the first drying stage is a known initial value and the calculated outlet air moisture is the new inlet moisture content for the second drying stage. Knowing the inlet air temperature and other process parameters in the second stage, the wet bulb temperature and drying time can be determined from Equations (3)-(5) and (10). Then the same actions as in the second stage are repeated for the remaining stages. The air mass flow through the multi-stage drying system is constant and the values for parameters y1-yn are calculated as follows:
u2dm τ
)ρM
(12a) dm22
da2a2u1
dm11
dmda1a1
ε(1Zρv
yτ)ρε(1Z
Mρvy
−=
−
&&21
:
undmnn
dmdanann1-un
dm1-n1
dm1-dan1-ann τ
)ρε(1ZMρv
yτ)ρε(1Z
Mρvy
−=
−−
&&1 (12b)
y1+y2+…yn = 1 (12c) It is important to note that sharing the fuel flow is only a computational action, and there is no reason to share the fuel flow in a real dryer. Since the values for parameters y1-yn are nown, approximative fuel moisture contents between drying chambers may be calculated
follows:
-y1(uin-uout) (13a) 2out 2in-y2(uin-uout) , u2in = u1out (13b)
: un,out = un,,inyn(uin-uout), un,in = un-1,out (13c)
in
ractice may be difficult to carry out.
kas
u1out = u1inu = u
Basically, the fuel flow can be shared in a real dryer, too. However, sharing the fuel flowp
30
igure 11. Sharing the fuel flow between drying stages in n-stage drying.
.3
he following conclusions can be drawn concerning the design of the dryer:
• In the design of the conveyor/fixed bed dryer, the air velocity must be slightly lower than the minimum fluidized velocity and the bed thickness must be at least so high that the drying air reaches its saturation point at the beginning of the drying.
alculation equations for determining the dryer dimensions in single- and multi-stage r dimensions is based on the use of the
F
3 Conclusions on the dryer design T
• Increasing the bed height still decreases the dryer size but the influence on dimensions is not as significant as for thin beds.
Cdrying are also presented. The determination of dryewet bulb temperature and fuel flow sharing.
Drying Drying Dryingstage 1 stage 2 stage n
y1Mdm , uout y2Mdm , uout ynMdm , uout
Δu = uin-uout
y1Mdm , uin y2Mdm , uin ynMdm , uin
Mdm , uin
y1 + y2 + ..+ yn =1
31
4 Integration and operation of the biofuel dryer in an industrial CHP plant
.1 Industrial CHP plant
igure 12 shows the energy and material flows in and out of the control region of a typical HP plant with a fluidized bed boiler in an integrated pulp and paper mill. In addition to e fluidized bed boiler, there are other boilers (e.g. a black liquor boiler, a gas turbine +
eat recovery boiler, and an oil boiler for peak load situations) at the mill, too. The
If the mill ine + heat recovery boiler may be found at the
ill. To discuss the questions as they are raised in this thesis, it is enough that we focus nly on the CHP plant shown in Figure 12.
t is defined as follows:
4 FCthhstructure of the entire energy generation system depends on the products the mill produces. For example, there is always a black liquor boiler if the mill produces chemical pulp.
produces mechanical pulp, a gas turbmo The total efficiency of the CHP plan
f
h
QPη Φ+
= , (14)
where P is the power generation, Φh the process heat generated in the form of steam, and Qf the fuel input into the fluidized bed boiler. The power-to-process heat ratio in CHP production is defined as follows :
h
Pα = Φ
(15)
he power-to-process heat ratio and efficiency are known parameters for each CHP plant. The fuel consumption c heat consumption is known.
T
an be calculated from equations (14) and (15) as the
32
Electricity to the mill
Extraction steam to the mill
Back pressure steam to the mill
Secondary heat from the mill
Figure 12. Energy and material flows in and out of a CH -plant at a pulp and paper mill. The main purpose of the CHP plant is to produce enough steam to cover the heat consumption of the mill at all times. The steam consumption at the mill is not constant but varies constantly depending on the production volume of the mill, the weather and theseason. Variation in steam generation means that the fuel input into the boiler varies, too. Usually, fuels are consumed in the following order: first, black liquor (if chemical pulp is
roduced), then bark from the debarking plant, and finally purchased fuels. Purchased fuels an be both bio and fossil fuels, and the CHP plant can use more than one purchased fuel.
e regoing order. For example, the availability of purchased fuels, operation hours of the
ebarking plant, and problems caused by the high moisture content of the bark may affect
al fuel moisture content be? 2. What heat sources will be used for the heating of drying air: only secondary heat,
secondary heat + steam or only steam? 3. What will the optimal number of drying stages be?
pcHowever, it is necessary to mention that the fuel consumption does not always follow thfodthe order of fuel consumption.
4.2 Integration of the dryer into the CHP process
4.2.1 Model description When the biofuel drying system discussed in this thesis is integrated into a CHP process, the following three questions will be raised:
1. What will the optimal fin
Bark from the debarking plant
Purchased fossil fuels (e.g. coal, oil, and peat)
Energy and material losses
Fluidised bed boiler
Barkdryer
G
Feed water tank
Condensate from the mill
make-up water
Back pressureturbine
Steam to thefeed water tank
Steam to the feed water pre-heater
Purchased biofuels (e.g. bark, forest residues,and saw dust )
Air
33
An optimization/calculation model has been created for analyzing quehas been developed using the calculation principles presented in chapter 3.2.
stions 1-3. The model
he objective function in the model is the net present value (NPV), which takes into
Taccount all cash flows over the economic lifetime of the dryer. Taking investment costs and maintenance costs as negative cash flows, the net present value for the dryer becomes [48]:
investmentC)i(
NPV+∑
kenanceintmaop )CNI(
−−
=τ=τ
τ0 1
, (16) =τ
he number of years over which
ss flow or a
wc
re NI is net income of drying in a time unit, k the totalash flows occur, and τop the annual operating hours of the dryer. Investment costs are
expressed as a function of a suitable capacity factor. For a continuous conveyor dryer operating as a cross-flow concept, a suitable capacity factor is the air maparameter which can be derived from the air mass flow (e.g. cross-sectional area or volume flow). If the dryer uses secondary heat, backpressure and extraction steam, the net income (NI) of drying in a time unit is calculated as follows:
ebsbsmfbsbs bαb
)α−+
+Φ
Φ
efanda
daeesesmf
eseses bηρΔpm
bαbη
)α(
&−+
+Φ
ΦΦ
where where M
bsshshmfvoutindm η
(bbl)uu(M NI & −−−=
ΦΦ
, (17)
heat consumption of the dryer, bsh e price of secondary heat, Φbs the backpressure steam consumption of the dryer, αbs the
ower-to-heat ratio in backpressure steam generation, η the efficiency of the CHP plant calculated from the fuel input, be the price of electricity, Φes the extraction steam consumption of the dryer, αes the power-to-heat ratio in backpressure steam generation, Δp the pressure drop over the drying system, and ηfan the efficiency of the fan. Equation (17) assumes that the fuel input into the boiler is adjusted on the basis of the heat consum tion of the mill and no extra heat is produced for increasing the power generation.
his is how CHP plants usually operate at pulp and paper mills. Because drying improves
generation or both. In some cases, a sufficient boiler capacity can be attained by installing a
dm is the dry mass flow of the fuel through the dryer, uin the initial fuel moisture, uout the final fuel moisture, lv the vaporization heat of water at a temperature of 25oC, bmf the price of marginal fuel, Φsh the secondarythp
pTthe effective heating value of the biofuel, marginal fuel can be replaced with biofuel. Savings in marginal fuel consumption (the first term in Equation (17)) represent a positive cash flow for a company. If steam is used as drying energy primary heat consumption increases and fuel input into the boiler may increase, too. In this case, earnings stem from the increased power generation. Depending on the initial values used the model calculates whether the earnings stem from the decreased marginal fuel consumption, increased power
34
dryer before the existing boiler, and a boiler investment is avoided. In these cases, savings would result from the avoided boiler investment. However, these cases are beyond the
ope of this thesis.
lation.
The drying system is assumed to consist of a conveyor, heat exchangers, air ducts, a function of air mass flow or cross-sectional area of
e dryer (also dependent on air mass flow) are shown in Table 3. Cost functions are based
sc Substituting (17) in (16) and expressing investment costs as a function of air mass flow, the objective function is completed. The aim of the model is to determine the final bark moisture content, the combination of heat sources, and the number of drying stages, which maximize the objective function. The entire model with boundary conditions is presented in more detail in Appendix paper III and is not discussed any further here. With regard to the application of the model, can be stated that the model calculates the drying temperatures, heat consumptions, and air mass flow maximizing the NPV in a case where the final fuel moisture content uout and the number of drying stages are given. To determine the final fuel moisture maximizing the NPV, the calculation must be made for different values of uout. On the basis of these values, the NPV can be drawn as a function of uout, and the optimal final fuel moisture content is found graphically (see Figure 13).
4.2.2 Results and discussion The model has been applied in a case study to analyze questions 1-3 mentioned at the beginning of chapter 4.2.1. In the case study, a virtual dryer is integrated into an existing CHP plant, which is situated at a pulp and paper mill in Finland. The material to be dried is soft wood (pine) bark. Experimentally determined drying curves for the bark are shown in Figure 10, and the same curves have been used in the calculation. The process values of the CHP plant, the temperature level of secondary heat, the initial bark moisture content and the dry mass flow of bark are based on data obtained from the mill where the CHP plant is situated. Prices of marginal fuel and electricity correspond to the price level in 2005. Table 2 shows a summary of all the initial values used in the calcu
covering and a fan. Cost functions as athon data obtained from equipment suppliers and reference [48]. To take into account costs such as instrumentation, lagging etc., the sum of the cost functions is multiplied by a factor of 1.6, which is based on information obtained from [48]. The determination of cost functions is explained in more detail in Appendix paper II. Maintenance costs are assumed to be 3 per cent of investment costs.
35
Table 2. Initial values for calculation cases. Temperatures in parentheses refer to Fig. 7a. Parameter Value Highest drying temperature when secondary h oeat is used (t2,t6) 70 C Highest drying temperature when backpressure steam is used (t3, t7) 120oC
ighest drying temperature when extraction steam is used (t4 , t8) 150oC aximum number of drying stages 3 nnual operating time of dryer 8400h ir velocity per free-sectional area of dryer 0.65m/s ower-to-heat ratio in extraction steam production 0.271 ower-to-heat ratio in backpressure steam production 0.336 fficiency of CHP-plant (equation (1)) 0.87 ressure drop of each drying stage (estimated) 400Pa ed thickness 200mm itial bark moisture content 1.5kg/kgdmulk density of dry bark 85kg/m3
/MWh MWh
kgda
HMAAPPEPBInBDry mass flow of bark 4kg/s
ctricity /MWh Price of ele 30euroroPrice of secondary heat
10.5eu
Price of marginal fuel 14euro/Economic life of dryer 10years
Interest rate 5%oOutdoor temperature (t1)
Outdoor humidity (x ) 5 C 0.004kg1 /
Table 3. Cost functions for main components of the dryer
Capacity ter Y
Additional pa r Equipment Relationship parame
ramete
Conveyor 2700Y Cross-sectional area
Air-water heat 9exchanger
ΔtY0.9 Air mass flow Δt is temperat se in heat exchanger
18ΔtY0.9 Air mass flow Δt is temperatexchanger
3770Y0.5 Air mass flow 0.9ΔpY0.7 Air mass flow Δp is pressure rying stage
200Y0.5 Cross-sectional area
ure increa
Air-steam heat exchanger
ure increase in heat
Air duct Fan drop of dCovering 1
In a basic case, the optimal drying system is determined using the initia
ying systems in a bl values shown in
alculation - tage dr asic case are shown 13.
the len he eco e and tricity and
heat on the optimal drying system is evaluated through the sensibility study. In study rame ice and other
s have the value 2. T ing values are used in the ity study calc leng ic lif inal fuel
/MWh, the elect ro/M h and the price of secondary heat 0-3euro/MWh. The price of secondary heat depends on the mill’s pricing policy. Some mills
Table 2. C results for 1 , 2- and 3-sin Figure
The effect ofsecondary
gth of t nomic lifetim prices of marginal fuel, elec
the sensibilityter
, one pa ter (e.g. the pr of marginal fuel) is changedhe followparame
ibil same as in Table
sens ulations: th of econom e 1-20 years, the price of marg6-20euro price of ricity 20-70eu W
36
include some costs (e.g pumping costs) in the price of secondary heat and others do not rice secondary heat at all. The maximum number of drying stages can be 3, and drying
e 13 but only the best case is shown in Figures 14-17. If the drying mperature is higher than 70oC, steam must be used for the heating of drying air in
igure 1sults at global optimum maximizing the net present value in a basic case (see initial
alues from Table 2).
psystems with more drying stages have not been considered in this study. Figures 14-17 show the results of the sensibility study. The results are calculated in a way similar to that shown in Figurteaddition to secondary heat. However, in some cases it may be more profitable to use only steam, not secondary heat, in the dryer. These cases are marked separately in Figures 14-17. 1000
1500
2000
2500
3000
3500
4000
4500
NP
V (k
euro
)
Initial bark moisture 1.5kg/kgdmPrice of secondary heat 0.5euro/MWhPrice of marginal fuel 14euro/MWhPrice of electricity 30euro/MWhEconomic lifetime of dryer 10yearsInterest 5%
1-stage drying
3-stage drying
2-stage drying
0
500
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
Final bark moisture [kg/kgdm]
Calculation results at global optimum
Number
of drying stages
Optimal drying
temperature [oC]
Optimal final bark
moisture [kg/kgdm]
Specific heat
consumption[kJ/kgH2O]
Investmentcosts
[keuro]
Maximum net present
value [keuro]
1 70 0.3 5860 4300 4053 2 70 0.3 4360 4520 4185 3 70 0.3 3890 4725 4071
F 3. Net present values as a function of final bark moisture contet and calculation rev
37
0
0.8
1
1.2
1.4
1.6
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Economic lifetime of dryer [year]
moi
stur
e [k
g/kg
dm]
0
2000
6000
8000
10000
12000
NP
V [k
€]
Initial bark moisture 1.5kg/kgdmPrice of secondary heat 0.5€/MWhPrice of marginal fuel 14€/MWhPrice of electricity 30€/MWhInterest 5%
70°C = optimal drying temperature
1 = optimal number of drying stages
0.2
0.4
0.6
Fina
l bar
k
4000
70°C1
70°C1
70°C1
70°C2
70°C2
70°C2
70°C2
70°C2
70°C
270°C2
70°C2
70°C2
70°C2
70°C2
70°C2
70°C2
Figure 14. Results of the sensibility study as a function of economic lifetime. Reading xample: If the economic lifetime is 5 years (see other initial values from the figure), the
figure shows that the optimal final bark moisture is 0.6 (columns), the optimal number of
price is 16 €/MWh (see other initial values from the figure), e figure shows that the optimal final bark moisture is 0.2 (columns), the optimal number
of drying stages 2, the optimal drying temperature 70oC and the net present value approximately 6000k€ (line).
e
drying stages 1, the optimal drying temperature 70oC and the net present value approximately 500k€ (line).
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Fina
l bar
k m
oist
ure
[kg/
kgdm
]
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
NP
V [k
€]
150°C1
150°C1
133°C1
70°C = optimal drying temperature2 = optimal number of drying stages
70°C2
70°C2
70°C2
70°C2
Initial bark moisture 1.5kg/kgdmPrice of secondary heat 0.5€/MWhPrice of electricity 30€/MWhEconomic lifetime of dryer 10yearsInterest 5%
Figure 15. Results of the sensibility study as a function of the marginal fuel price. Reading example: If the marginal fuel
6 8 10 12 14 16 18 20
Price of marginal fuel [€/MWh]Seconadry heat consumption is 0MW
th
38
00k€ (line).
0
0.1
0.2
0.3
0.4
0.5
15 20 25 30 35 40 45 50 55 60 65 70
Price of electricity [€/MWh]
Fina
l bar
k m
oist
ure
[kg/
kgdm
]
0
2000
4000
6000
8000
10000
12000
NP
V [k
€]
70°C2
70°C2
70°C2
70°C2
70°C2
70°C2
70°C2
150°C = optimal drying temperature1 = optimal number of drying stages
150°C1
150°C1
150°C1
120°C1
Seconadry heat consumption is 0MW
Initial bark moisture 1.5kg/kgdmPrice of secondary heat 0.5€/MWhPrice of marginal fuel 14€/MWhEconomic lifetime of dryer 10yearsInterest 5%
Figure 16. Results of the sensibility study as a function of the electricity price. Reading example: If the electricity price is 40 €/MWh (see other initial values from the figure), the figure shows that the optimal final bark moisture is 0.3 (columns), the optimal number of drying stages 2, the optimal drying temperature 70oC and the net present value approximately 4000k€ (line).
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.5 1 1.5 2 2.5 3
Price of secondary heat [€/MWh]
Fina
l bar
k m
oist
ure
[kg/
kgdm
]
0
1000
2000
3000
4000
5000
6000
NP
V [k
€]
70°C1
70°C2
70°C = optimal drying temperature2 = optimal number of drying stages
70°C3
70°C3
70°C3
70°C3
Initial bark moisture 1.5kg/kgdmPrice of marginal fuel 14€/MWhPrice of electricity 30€/MWhEconomic lifetime of dryer 10yearsInterest 5%
Figure 17. Results of the sensibility study as a function of the secondary heat price. Reading example: If the secondary heat price is 1.5 €/MWh (see other initial values from the figure), the figure shows that the optimal final bark moisture is 0.4 (columns), the optimal number of drying stages 3, the optimal drying temperature 70oC and the net present value approximately 27
39
Using
condary heat for the heating of drying air, consists of two drying stages and dries the el/bark flow to the final moisture content, which is in the order of 0.3kg/kgdm (see Figure
3). The reasons why the results in Figures 14-17 behave as they do are explained in ppendix paper III.
rom the viewpoint of the dryer investment the most interesting question is how fuel and lectricity prices will develop in the future.
able 4 shows how the average market prices of the most common mill fuels have eveloped in Finland between 2001-2005. Table 4 proves the fact that fuel prices have creased considerably in 2005 due to the introduction of the emissions trading (see prices cluding price of CO2-ton). Table 4 also reveals that the price of natural gas has increased
lmost 20 percent from June 2004 to September 2005, even though the effect of emissions ading is excluded. Natural gas is the most important fossil mill fuel used by the forest dustry in Finland [5]. Even though it is difficult to predict the price behavior of CO2-ton
t some mills, biofuel may be the marginal fuel part of the year or the whole year, and one ay ask whether drying is profitable in these cases. Because of the emissions trading veral municipal energy companies have started to replace fossil fuels with biofuels, for
igure 16 shows that the electricity price has no influence on the profitability and design
ween 23 and 36 euro/MWh in the Nordic lectricity markets during the years 2001-2005 [49]. In this kind of calculation, it is more
oes not have ny significant influence on the design parameters of the optimal drying system and the
the initial values shown in Tables 2 and 3, the optimally designed dryer uses only sefu1A Fe Tdininatrinmerely on the basis of year 2005, it is probable that fuel prices will not decrease considerably in the near future. On the contrary, prices may still increase. If the expected behavior of fuel prices continues, the profitability of the dryer investment improves and the optimal drying system is similar to the one in our basic case (see Figure 13). Amseexample, in district heating, and the market price of biofuel has increased. From the viewpoint of the forest industry, this means that excess biofuel can be sold outside the mill, and drying may still be profitable even if biofuel is the marginal fuel for most of the year. To analyze these cases correctly, the market price of biofuel must be used as a fuel price. Fparameters of the optimal drying system until the price exceeds 45 euro/MWh. The annual average price of electricity has varied betereasonable to use the average electricity price over a longer period rather than instantaneous prices. The Nordic electricity markets are characterized by the following factors: (i) Circa fifty per cent of electricity is produced by hydropower and the price usually decreases and increases during rainy and non-rainy years, respectively. (ii) Long cold periods in the winter usually increase the consumption and the price. (iii) Instantaneous electricity prices may be very high due to the reasons mentioned in points (i) and (ii). (iv) Over thirty percent of electricity is produced by nuclear power and renewable energy sources, which are CO2-free energy production forms. (v) A new nuclear reactor will be introduced in Finland in 2010 and its effect on the future electricity price is unclear. For these reasons it is difficult to estimate the behavior of the average electricity price in the future. On the basis of the average electricity prices during the years 2001-2005, the electricity price da
40
profitability of drying. If the average electricity price reaches permanently a clearly higher level in the future, it may have a more important role in the design of the optimal drying system. Finally, it is necessary to state that power-to-heat ratios used in the calculation are relatively high for industrial CHP plants. If the power-to-heat ratios were lower, the electricity price in Figure 16 should be even higher before the optimal design values and the profitability would change. The values used for economic lifetime and the price of secondary heat are decided by the
2-
mill and can vary depending on the actor. On the other hand, these parameters seem merely to influence the final bark moisture content and the number of drying stages. Table 5. Development of fossil fuel prices in Finland between 2001 and 2005 [50]
Fuel prices
Fuel prices including price of COton
Coal Natural gas
Peat Heavy fuel oil
Price of
CO2-ton
Coal Natural gas
Peat Hefue
[€/MWh] [€/MWh] [€/MWh] [€/MWh] [€/t] [€/MWh] [€/MWh] [€/MWh] [€/M6/2001 12.9 15.8 9.2 25.0
avy l oil Wh]
12/2001 12.4 15.7 9.2 21.8 6/2002 11.6 15.2 9.2 24.3 12/2002 11.3 15.7 9.2 25.5 6/2003 11.0 16.1 9.3 24 12/2003 11.6 15.6 9.4 23.3 6/2004 13.4 15.3 9.4 26.2 12/2004 13.4 16.1 9.7 24.5 3/2005 13.8 16.2 9.8 29.5 10 17.2 18.2 13.6 36/2005 13.3 17.0 9.8 33.5 23 21.1 21.6 18.6 39/2005 13.4 18.2 8.2 39.8 23 21.2 22.8 16.9 4Emission factors for fuels based on lower heating values from IPCC [kg CO
2.3 9.9 6.2
2/MWh fuel used] : Coal 340.28, Natural gas 201.80, Peat 381.26, Heavy fuel oil 278.26
4.3 Operation of the dryer in an industrial CHP plant
.3.1 Model description 4 Th cha s simulation results of the mis pter present odel that have been developed for analyzing the operation of the dryer integrated into an industrial CHP plant at a pulp and paper mill. The ma tiv he an is t uat th eas e tequip the dryer with a rol system or n
n ly c rable variation in the fuel moisture content. The variation is ep t on ason instan us weather. Season and weather also affect erature and humidity of the drying air before the first heat exchanger. The created
on l ca es th tgoing fuel m isture content on the basi of the el re c t, am t temp e and humidity.
in objec e of t alysis o eval e whe er it is r onabl o cont ot.
There is ormal onsidemainly d enden the se and taneothe tempsimulati mode lculat e final/ou o s initial fu moistu onten bien eratur
41
The availab f pr com ed the model-based contpli kno e-bas odels model uilt for p ocess sim lation use pro contr 1]. In conn n, mulation mod alsoto odel-based control of the dryer. The pose e m s to inemp re, o ing te rature e cas mul e d in s way
ost del has been created for a
rying system similar to one shown in Figure 7a.
output/controlled pulated variables
nt temperature, humidity) cannot be adjusted by a control (e.g. drying temperature, dry mass flow of fuel) can be
djusted either manually or automatically. In the model, all variables that are possible to
able 6. Fixed, load, controlled and manipulated variables in the simulation model
ility o ocess puters has advanc rol with the use of either ex cit or wledg ed m , and the same b r ucan be d for cess ol [5 this ectio the si el is applied the m pur of th odel i determ drying te eratu r dry mpe s in th e of ti-stag rying, uch a that the desired final fuel moisture content is always achieved at minimum operating c(i.e. by maximizing net income/earnings from drying). The mod Generally, drying process variables can be divided into input and variables [52]. Input variables can be further classified into load and mani[52]. Load variables (e.g. ambiesystem. Manipulated variables amanipulate are not adjusted, but have a fixed value, which is the same as the design value. Table 6 shows the group to which each variable belongs in this model. If the variable that is possible to manipulate is not adjusted, it is called a fixed parameter in Table 6. TParameter/variable Symbol Group of variable Explanation Air mass flow
dam& Fixed Same as the design value
Particle size - Fixed Same as the design value Material properties of bark - Fixed Same as the design values Cross-sectional area of the A Fixed Same as tdryer
he design value
Bed height Z Fixed Same as the design value Dry mass flow of bark
dmM& Fixed Same as the design value
Initial bark moisture content
uin Load Depends on material, weather and season
Outdoor humidity x1 Load Depends on weather and season Outdoor temperature t1 Load Depends on weather and season Final bark moisture content uout Controlled Same as the desired set value Drying temperatures t4, t8, t12….. Manipulated Model calculates optimal drying
temperature(s) Heat inputs Φ ,Φ ,Φ Manipulated Drying temperatures are adjusted to sh bs es
optimal ones by manipulating heat inputs
Table 6 shows that fuel and air input (dry mass flows of fuel and air) into the dryer are assumed to be fixed and manipulated parameters are drying temperatures, which are determined in each drying stage by the model. To be precisely, the actual parameters to be manipulated are heat inputs in to the dryer (also shown as a manipulated parameter in Table ), which are adjusted by conventional controllers in order to achieve calculated drying 6
temperatures. Because the model calculates drying temperatures, they are meters i f heat i
called manipulated para nstead o nputs.
42
The simulation model is esented in de Appendix cussed any h odel to model-based contro ure content
nown on-line. A conceivable approxim discussed in the next chapter and the resented in
or on-line determ on of init nt
way to the mois f fuel At least two samples are ta the fuel flow and kept in an oven
24 hbe assumed the samples are completely dry and the initi
et basis is defined as follows:
pr tail in paper IV and is not disfurther here. To apply tmust
e m l the initial fuel moistbe k method to ately determine the initial fuel
moisture on-line is chapter 4.3.2.
simulation results are p
4.3.2 Method f inati ial fuel moisture conte The most reliable determine initial fuel ture content is the drying osamples in an oven. at 105±2
ken from longer thanoC for at least 16 hours but not ours [53]. After the drying, it can
al moisture content calculated on w
1
21o m
mmw −= , (18)
where m1 is the mass of the sample before drying and m2 after drying. Although the method is reliable and simple, it is time consuming and it must be carried out manually, which is not desirable. Appendix paper V presents a simple method for approximately determining the initial fuel moisture content. The method is based on the on-line measurement of the outlet air moisture at the beginning of drying. The idea of the method is to find a correlation between fuel moisture content and the change in the outlet air moisture content. The idea was experimentally tested by drying regular shaped wood particles (20x20x5mm) in a test rig (see Figure 3). The evaporation surface in each test was the same but the moisture contents f the particles were different. Figure 18 shows the change in the outlet air moisture for
n
odifferent initial moisture contents of particles and also for particles which were frozebefore drying.
43
10.0
12.0
14.0
16.0
18.0
istu
re [g
/kg d
a]
61.4% w.b.
56.0% w.b.
Inlet air temperature 70°C, Air velocity 0.6m/sInlet air moisture 0g/kgda
0.0
2.0
4.0
6.0
8.0
0 100 200 300 400 500 600 700
Time [s]
Out
let a
ir m
o
54.4% w.b.
47.0% w.b.= initialsample moisture
Frozen particles
Figure 18. Change in outlet air moisture content for different initial moisture contents of mple particles and for frozen particles.
igure 18 shows that there is a relatively clear correlation between the change in outlet air oisture content and the initial moisture content of the particles. Even a relatively small
IVt but
om the
of the determ than
faster To
determ
sa Fmmoisture difference between particles (56% and 54.4 %) can be observed. Appendix paper
proposes that outlet air moisture content is measured in the drying chamber after the fuel has entered the chamber. This is one possible way to carry out the measuremenprobably not the best way. A more comprehensive way is to take a small sample frfuel flow to a separate test chamber before the drying chamber. By measuring the outlet air moisture content in a separate chamber with constant volume the evaporation surface is approximately the same in each measurement, which improves the accuracy
ination. In addition, the bed height in a separate chamber can be clearly lowerin the drying chamber, in which case the change of outlet air moisture content is observed
.
ensure the suitability of the method in the case of a commercial dryer, measurements for actual biofuels with varying particle size must be done. Instead of outlet air moisture content, measurement of outlet air temperature may give sufficient information for the
ination of initial fuel moisture content, and should be experimentally tested, too.
44
4.3.3 Results and discussion The model has been applied to a biofuel drying study where a virtual dryer is integrated into an actual CHP plant. The process values of the CHP plant are the same as in the case study in Chapter 4.2 (see Table 2). Chapter 4.2 concludes that the optimally designed dryer uses only secondary heat as heating energy when it operates at its design values. Table 7 shows design values for four dryers which are designed to use secondary heat as a heat source. In simulations, these dryers are first equipped with a model-based control and then the results are compared with similar dryers with no control. The material to be dried is soft wood (pine) bark. The load variables (bark moisture contents, outdoor temperatures, and humidity values) for the drying simulations are
of 000 to the end of 2003 [49]. The prices of marginal fuel and secondary heat are 10€/MWh nd 0.5 €/MWh, respectively.
able 7. Design values of dryers in simulation cases
[kg/kgdm]
rk e content
[kg/kgdm]
Number of drying stages
Drying temperature [oC]
Residence time [s]
obtained from the same mill where the CHP-plant is situated (see load variables in Appendix paper V). In the simulations, the load variables have 122 values and they cover the whole calendar year (load values change every 72 hours). In practice, the inlet values may change more frequently. The electricity prices used in the calculations change every month. The prices are based on data from the Nordpool for the period the beginning 2a TCase Initial bark
moisture content Final bamoistur
1 1.53 1.00 1 70 1035 2 1.53 0.667 1 70 1695 3 1.53 0.333 1 70 2730 7 1.53 0.667 3 70 (in each stage) 1895 Design values similar in all cases: Initial bark moisture content 1.53 kg/kgdm, bed thickness 200mm, air velper free sectional area 0.65m/s, outdoor temperature 5
ocity heat nger
oC, humidity 0.004kg/kgda, , effectiveness for secondaryexchanger 0.89, effectiveness for backpressure steam exchanger 0.73, effectiveness for extraction steam excha0.67 The results for the simulation cases are shown in Figures 19-20. The figures show final bark moisture contents and annual net incomes of drying for the dryers equipped with the model-based control and for the dryers with no control. Net incomes of drying are expressed as euro/tdm. The simulation results are independent of the dry mass flow and the net incomes are directly proportional to dry mass dried in the dryer. Appendix paper V also discusses cases where the dryer is designed to use both secondary heat and backpressure steam as heat sources. On the basis of our results in Chapter 4.2.2, these cases are not relevant.
45
ent and net income of drying in cases 1-3 l r with control b)
dryer without control.
Figure 19. Behavior of final bark moisture cont(see cases from Table 7) on the basis of simulation calcu ations: a) drye
Figure 20. Behavior of final bark moisture content and net income of drying in case 4 (see case from Table 7) on the basis of simulation calculations: a) dryer with control b) dryer without control (Note case 4 is case 7 in Appendix paper V).
Case1 Case2 Case3 Average final moisture content, kg/kgdm
0.971 0.675 0.344
Standard dm
0.222 0.181 0.107 deviation, kg/kgNet income, €/tdm 3.325 5.024 6.792
Case1 Case2 Case3 Average final moisture content, kg/kgdm
0.992 0.666 0.330
Standard deviation, kg/kg
0.046 0.008 0.000 dm
Net income, €/tdm 2.746 4.442 6.004
a) Cases 1-3, ith control2.50
Case 2
Case 1
Initial bark moistureAverageDryer w 1.53kg/kgdm
deviation 0.25 kg/Standard kgdm
1.00
1.50
2.00
ture
[kg/
kgdm
]
0.00
0.50
0 1000 2000 3000 4000 5000 6000 7000 8000
Operating time [h]
Ba
Case 3
rk m
ois
b) Cases 1-3, Dryer without control
0.00
0.50
1.00
0 1000 2000 3000 4000 5000 6000 7000 8000Operating Time [h]
Bar
k m
oist
ure
[kg/
kgdm
]
1.50
2.00
2.50 Initial bark moistureAverage 1.53kg/kgdm
Case 3
Case 2
Standard deviation 0.25 kg/kgdm
Case 1
Dryer with
Dryer without
control control Average fina
Case 4, Bark moistures for a dryer with and without control
0.00
0.50
1.00
1.50
2.00
2.50
0 1000 2000 3000 4000 5000 6000 7000 8000
Operationg time [h]
Bar
k m
oist
ure
[kg/
kgdm
]
Initial bark moistureAverage 1.53kg/kgdm
Standard deviation 0.25 kg/kgdm
Dryer without control
Dryer with control
l
t, moisture contenkg/kgdm
0.666 0.642
Standard deviation, kg/kgdm
0.008 0.129
Net income, €/tdm
4.784 5.348
46
The simulation results show that it is possible to stabilize the nal bark moisture content by controlling the drying temperatures. The results also show that the deviation in final bark moisture content decreases even when the dryer has no control. The lower the designed value for the final bark moisture content, the more the deviation decreases. The net income from the drying increases, when the dryer has no control. are two reasons for bigger the net income when the dryer has no control. Firstly, the whole potential of affordable secondary heat is always used in drying. In the case of the control, the whole potential is not used if the initial fuel moisture content w the sign value. Secondly, the dryer never uses more expensive steam in drying. In the case of control steam is consumed, if the initial fuel moisture content is above the design value. On the basis of the assumptions made in the case study, the need for a control cannot be justified on economic grounds. However, there may be a need for a control, if the use of homogenous, and not only dry, fuel has any notable influence on the following issues:
• Predictive control of the power plant in cases where the heat consumption at
n analysis of the above mentioned issues is beyond the scope of this study but should be
l in the boiler, the ryer may need control. On the other hand, mixing of moist fuel with dry fuel after the
dryer may a e of too dry fuel in the boiler. Otherwise the need for control depends on analysis results of the previously mentioned issues. In the case of a n fuel with a constant heating value) if the dryer is equipped with a control for some reason. The model remodel-based c
fi
There
is belo de
• Operation and control of the boiler • Boiler efficiency
the mill changes. • Deviations in (live or process) steam pressure resulting from changes in heat
consumption at the mill. Ainvestigated to be able to answer ultimately the question about the need to control the final fuel moisture content. If the use of homogenous fuel has positive influences on some of the above mentioned issues, the control may have economic grounds, too. The answer is also slightly dependent on whether the dryer is integrated into an existing or a new boiler. Existing fluidised bed boilers have been designed for moist and heterogeneous fuels. In particular, bubbling fluidised bed boilers (BFB boiler) may not even tolerate too dry fuel, which can cause, for example, ash melting as a result of the excessively high combustion temperature. To avoid the use of too dry fued
be n alternative method to avoid the us
ew boiler, the boiler can be already designed for dry and homogenous fuel (i.e.
p sented in Appendix paper V is a conceivable method of carrying out the ontrol of the dryer.
47
4.4 Conclusions on integration and operation of the dryer The following main conclusions can be drawn concerning integration and operation of the dryer at industrial CHP plant:
• Using current energy prices and an economic lifetime of 10 years with an interest rate of 5%, the optimally designed dryer uses only secondary heat for the heating of drying air, consists of two drying stages and dries the fuel/bark flow to a final moisture content which is in the order of 0.3kg/kgdm (see Figure 13).
• If fuel prices are expected to rise further in the future, the profitability of the dryer investment improves and the optimal drying system is similar to the one in our basic case (see the first bullet).
• The electricity price has not any significant influence on the design parameters of the optimal drying system and the profitability of drying in the case of study until the annual average price exceeds 45-50 euro/MWh in the case study (see Figure 16).
• Predicting future electricity prices is difficult in the Nordic electricity markets, and the rule of thumb is that expected earnings from drying primarily stem from
e are no
mogenous fuel may have some positive influences on the
decreased marginal fuel consumption, and not increased power generation. • On the basis of the assumptions made in the simulation case study, ther
economic grounds for homogenizing the final fuel moisture content by means of control.
• However, the use of hoboiler and the power plant process, which may favour the introduction of control. To be able to ultimately answer the question about the need to control the final fuel moisture content, these influences need to be analysed in collaboration with power plant specialists.
• If the dryer is equipped with control it can be carried out as a model-based control using the model presented in Appendix paper V.
48
5 Evaluation of energy efficiency in drying
he main goal of this chapter is to present two methods for evaluating the energy efficiency drying. The evaluation methods are the specific heat consumption and the irreversibility te. Both evaluation methods are applied to two flow sheets, which are alternative methods
f decreasing the heat consumption in drying. The flow sheets considered in this chapter re multi-stage drying (MSD) and single-stage drying with partial recycling of drying air SD). MSD is the primary method of improving the energy efficiency of drying in this esis. SSD in introduced as an alternative flow sheet for improving the energy efficiency. o further decrease the heat consumption, both drying systems are provided with an xhaust air heat recovery unit.
.1 Single- stage drying with partial recycling of drying air
igure 21a shows a continuous single-stage dryer with partial recycling of spent air cluding a heat recovery unit. Figure 21b shows also a multi-stage drying system with a
heat recovery unit. Multi-stage drying is discussed in more detail Chapter 3.1. The changes gle-stage drying with
t f the spent drying air
Tinraoa(SthTe
5 Fin
of the states of the drying air on an enthalpy humidity chart in sinar ial recycling of air is illustrated in Figure 22. Partial recycling op
and the heat recovery reduce heat consumption in drying process, because the air temperature before the heating increases. Because of the recycling, the fresh drying air has a lower potential for taking up moisture, which leads to a bigger air mass flow through the drying chamber.
Figure 21. Methods of improving energy efficiency a) Partial recycling of spent drying air+ heat recovery (SSD) b)Multi-stage drying + heat recovery (MSD).
Dryingchamber
Heating
Recycling air
Sup airply
Moist fuel Dry fuel
HRU
Exhaust airHRU=Heat recoveryunit
Dryingchamber
Dryingchamber
Dryingamberch
Supply air
Heating Heating Heating
Exhaust air
Moist fuel Dry fuel
HRU
a) b)HRU=Heat recoveryunit
49
Figure 22. Changes of the states of the drying air on an enthalpy humidity chart in single-stage drying with partial recycling of air.
5.2 Evaluation methods Traditionally, the most common way to evaluate the energy efficiency in drying is the specific heat consumption [54]:
e
n
ii
mq
&
∑== 1
Φ , (19)
here n is the number of drying stages, φ the heat input into the drying system, and the
lance and it does not pay any attention to the temperature level at which drying occurs. If the dryer is integrated into a power generation
t irrelevant from the perspective of energy efficiency. on method must be based on the
Heat recovery from
Mixture point
Translations: kuiv. ilm.= dry air torr luft = dry air kg vettä/ kg kuiv.ilm. = kg water / kg dry air kg vatten / kg toff luft = kg water / kg dry air
exhaust air
Heating period
Drying period
w em&evaporation rate of the dryer.
quation (19) is based on the energy baE
process the drying temperature is noo take into account the effect of temperature, the evaluatiT
Second Law of Thermodynamics. The term irreversibility rate is presented in [55] and is defined as follows:
50
] Ti ri=1
sm[TIm
in
iiio ∑∑
=
−=1
ΦΔ , (20)
where for each stream
tem is taken as positive. Tr is the temperature at the point on er is taking place [55]. Generally, the term
, (21)
exergy of the fuel, Ea the exergy of the streams leaving the control region, W the mechanical work, Id the
t the main aim in a CHP-production but the
tail in Appendix paper VI.
.3 Results and discussion
med to be 10oC higher than the let air temperature of the heat exchanger. In the heat recovery unit, the minimum
relative humidity of 60%. Mechanical work puts into the dryer and the heat losses are neglected. Neither are there any indirectly
To is the temperature of the surroundings, Δs the change of entropyof matter through the system, and Φi the heat inputs and outputs to the system. In this paper, the heat input to the sys
e boundary of the system where the heat transfthΦi/Ti is known as a thermal entropy flux [55]. If the steam in a turbine can expand to the pressure corresponding to the temperature of the surroundings, the exergy balance for a steam power plant with a dryer inside the control region is [55]: xdPaf IIWEEE ++=−+ where Ef is the xergy of the air entering the control region, Ep the eirreversibility rate of the drying and Ix the irreversibility rates of the other sub-processes. It can be assumed that Ep and Ix remain almost the same regardless of the drying process. The difference between the irreversibility rates of two drying processes gives how much mechanical work (electricity) will be lost if the dryer with the higher irreversibility rate is used instead of the dryer with the lower rate. However, it is necessary to state that maximizing the power generation is usually nomain aim is to produce heat for manufacturing processes. Determination of the heat consumption and irreversibility rate for MSD and SSD are explained in de
5 Equations (19) and (20) have been applied to a drying case for evaluating the energy efficiency of the drying process. In the drying case, moist biofuel is dried from an initial moisture content of 1.5kg/kgdm to a final moisture content of 0.3kg/kgdm, and the dry mass flow of the fuel is 1kgdm/s. Two alternative heat sources are available for the heating: steam at a pressure of 3bars (133oC), and water at a temperature of 80oC. The minimum temperature difference between air leaving the heat exchanger and the heat source entering the heat exchanger is 5oC, in which case the inlet air temperatures of the drying chambers are 128oC and 75oC. The outlet temperature of water is assuintemperature difference is 5oC. The outdoor temperature is 15oC and the absolute moisture content 0.0064kg/kgda corresponding to ain
51
supplied heat inputs into the drying tur ted after the drying chamber.
chamber. The drying air is assumed to be fully asa
Figure 23 shows heat consumptions and the irreversibility rates for multi-stage drying as a function of drying stages. Heat consumptions and the irreversibility rates for SSD are presented in Appendix paper VI. To compare the heat consumption and irreversibility rate between MSD and SSD, the dryer dimensions are altered so that they are the same. It is assumed that the dryer dimensions are proportional to the cross-sectional area of the dryer, which depends on the air mass flow as follows:
ada
a
vm
Aρ&= , (22)
The term am& is defined as the product of the number of drying stages and the mass flow of
air ss flow through the drying chamber is the same as & . Instead of the air mass flow or the cr nal area, the energy consumption and
dry air through the multi-stage dryer ( am& = damn & ). The mass flow am& is first determined for the MSD, because the number of drying stages must be an integer. In SSD, the recycle ratio is given such a value that the ma
damn oss-sectiothe irreversibility rate are expressed as a function of a dimensionless value, defined as follows:
aref
da
ref mm
nAA
&
&= , (23)
where A/Aref is the ratio of dryer areas and the mass flow of the dry air in single-ng w
temp
arefm&
stage-dryi ith no recycling ( R=0 and n=1). This is the smallest dryer, and the value of RDA is always bigger than 1 when the number of drying stages increases. In the example calculations, the maximum number of drying stages is reached when the recycle ratio is approximately 0.9. Figure 24 shows the comparison results for the inlet air temperature 75oC, in which case air is heated using secondary heat. Comparison results for the inlet air
erature 128oC are presented in Appendix paper VI.
52
Figure 23. Heat consumptions and irreversibility rates for multi stage drying system
=15oC). including heat recovery from exhaust air (To
igure 24. Comparison of energy consumption and irreversibility rate between single- age drying with recycling (SSD) and multi-stage drying (MSD) for an inlet temperature
of 75oC ( To=15oC ). Figure 23 shows that the drying temperature has little effect on heat consumption. The difference is in the order of 100kJ/kgH2O. In stead, the irreversibility rate clearly depends on the drying temperature. The differences between irreversibility rates are several hundred kilojoules per mass of water evaporated. When steam is used, heat transfer into the drying system occurs at a higher temperature level than in the case of secondary heat. The average temperature difference between the heat source and air is also higher for steam, because the steam temperature remains constant. Heat transfer occurring over a high temperature difference also increases the irreversibility rate. The difference between the irreversibility rates gives theoretical losses in electricity production, which is more relevant information than the difference between heat consumptions when the dryer is integrated into a power generation process. However, it is important to consider that the irreversibility rate gives
Fst
Inlet temperature 75°C
26
2720
2760
2800
2840
2880
2920
1 1.3 1.6 1.9
Hea
t con
sum
ptio
n [k
J/kg
H2O
]
80
A/Aref
SSDMSD
Inlet temperature 75°C
0
50
100
150
200
250
1 1.3 1.6 1.9
Irrev
ersi
bilit
y ra
te [k
J/kg
H2O
]
A/Aref
SSDMSD
Multi-stage-drying
20
26
27
27
28
28
28
29
0 5 10 15
Number of drying stages
Hea
t con
sum
ptio
n [k
J/kg
H2O
]
Multi-stage-drying
700
2O]
0
100
200
300
400
500
600
0 5 10 15
Number of drying stages
Irrev
ersi
bilit
y ra
te [k
J/kg
H80
40
75°C75°C00 128°C128°C
60
20
80
53
only a theoretical estimate of losses. In practice, it is difficult to utilise secondary heat in
w because the heat losses and the pressure drying air is assumed to be fully saturated is provided with a heat recovery unit. Only
erature differences in the heat exchangers have umptions and irreversibility rates are good gy efficiency of real air drying systems.
s with the same dimensions (Figure 24) shows e for both systems. The difference varies from 0
electricity production. The calculated values in Figure 23 are very lodrops of the drying system are neglected, the after the drying chamber, and the drying systemthe increase in fuel temperature and tempbeen considered. The calculated heat consreference values for the estimation of the ener The comparison between the drying systemthat the heat consumption is almost the samto 50kJ/kg . However, the irreversibility rate
tly more energy fficient drying system than single-stage drying with partial recycling of the drying air.
H2O is smaller for MSD than for SSD. The main reason for the lower irreversibility rate is the entropy change, which becomes smaller for MSD. According to these results, multi-stage drying seems to be a slighe
5.4 Conclusions on evaluation of energy efficiency
• By means of various flow sheets and heat recovery from exhaust gas the lowest possible heat consumptions in air drying are close to 2700 kJ/kgH2O.
• If the energy used in drying can be converted into mechanical work, the evaluation of energy efficiency should be based on exergy rather than on energy analyse.
• The irreversibility rate is a conceivable method for evaluating the energy efficiency of drying processes from the viewpoint of the Second Law.
• To decrease the value of the irreversibility rate, the drying temperature should be as low as possible, and the heat transfer should occur over small temperature differences.
• According to the results shown in Chapter 5.3, multi-stage drying seems to be a slightly more energy-efficient drying system than single-stage drying with partial recycling of spent drying air, when the dryer dimensions are the same.
54
6 Concluding remarks The tcase st ryer uses only secfuel co e in the ua basic owever, it is necessary to statecase anprofitability of the dryer investment are presented in this thesis. At the moment, the pro c On the not seem to
o economic grounds to control the final fuel moisture content by adjusting heat inputs into e dryer. However, the use of homogenous, and not only dry, fuel may have positive fluences on the operation/control of the boiler, boiler efficiency and predictive control of e power plant when heat consumption changes at the mill. To be able to answer
ltimately the question about the need to control the final fuel moisture content, these uestions should be analysed. If the dryer is equipped with control it can be carried out as a odel-based control using the model presented in Appendix paper V.
xperimental tests show that critical moisture content is high for woody-based fuels and it necessary to use diffusion-based drying models to determine drying the times eoretically in a fixed bed. Drying times calculated by the constant drying model are too accurate compared to actual drying times, even though the final fuel moisture content was
learly higher than the fiber saturation point. From viewpoint of both constant and iffusion-based drying models, the extremely wide particle size distribution in biofuel flow
also sets challenges for the development of theoretical drying models. In this study, drying ve been determined experimentally in a fixed bed drying times are analogous to a continuous cross-
n as an evaluation method is that it
in egration and operation of the dryer at an industrial CHP plant has been analyzed in a udy. According to the case study results, an optimally designed d
ondary heat, not steam, as a heat source, and earnings stem from decreased marginal nsumption, not increased power generation. If fuel prices are assumed to increas
fut re, as seems likely, the optimal design parameters of the dryer are similar to those in case and the profitability of the dryer investment improves. H that the profitability of the dryer investment depends always on the actor and the d must be assessed separately in each case. Guidelines for the assessment of the
spe ts for profitable dryer investments are good.
basis of the assumptions made in the simulation calculations, there does nthinthuqm Eisthincd
times used in the calculation models haatch dryer. Experimentally determinedb
flow process. Guidelines for dimensioning the conveyor/fixed bed dryer in the case of multi-stage drying have bee presented. In the design of the dryer, air velocity must be slightly lower than the minimum fluidized velocity and the bed thickness must be at least so high that the drying air reaches its saturation point at the beginning of the drying. Determining the dryer dimensions in the case of multi-stage drying is based on the use of the wet bulb temperature and fuel flow sharing. Specific heat consumption is the most common method of evaluating the energy efficiency n drying. The weakness of specific heat consumptioi
does not pay any attention to the drying temperature. On the basis of the calculated results the drying temperature has a quite small effect on the calculated heat consumptions. To take into account the effect of the drying temperature on the energy efficiency an alternative evaluation method, i.e. the irreversibility rate, has been introduced. The results
55
show that there is a clear difference between irreversibility rates depending on the drying temperature. If the energy used in drying can be converted into mechanical work, irreversibility rate is a more comprehensive method of comparing energy efficiency between different drying processes than specific heat consumption.
7 Recommendations for the future research and development of the dryer
Building a pilot-dryer is a necessary step before commercialization of the dryer. Robustness is one of the key requirements for a commercial dryer and it is impossible to evaluate the robustness reliably without experiences from a pilot-dryer. Fuel input and output, dust formation, blockage risk, fire risk, fouling, corrosion, the need for maintenance as well as moisture distributions in vertical and horizontal directions are the main issues that must be analysed by means of the pilot-dryer. A pilot-dryer is also the most reliable way to specify resented design parameters.
nalyse these issues. If the control of the final fuel moisture content mpared with the uncontrolled dryer, the idea of the model-based
p The use of dry and homogenous fuel may have positive influences on the operation and control of the boiler and power plant process. The main issues to be analysed in this connection are listed in Chapter 4.3.3 but have not been investigated any further in this study. Especially, from the viewpoint of controlling the final fuel moisture in the dryer it would be important to antails clear benefits coe
control must be implemented in the pilot-dryer. Both the boiler analysis and the possible implementation of the model-based control must be carried out in collaboration with boiler and control specialists.
56
References
soilla (Reduction of the carbon dioxide emissions by biomasses in the energy production of the forest industry). Publications of Laboaratory of Energy Economics and Power Plant Engineering, Raport A2. Espoo, Finland: Helsinki University of Technology; 1999.
] Finnish Forest Industries Federation. Finnish Forest Industries Yearbook 2005, Statistics 2004. Helsinki, Finland: Finnish Forest Industries Federation; 2005.
] Statistics Finland. Energiatilastot 2004 (Energy Statistics 2004). http://www.tilastokeskus.fi/til/ekul/index.html, February 28, 2006.
] Huhtinen M, Hotta A. Combustion of Bark. In: Gullichsen J, Fogelholm C-J, editors. Chemical Pulping. Helsinki, Finland: Papermaking Science and Technology; 1999. p. 203-301.
] Alakangas E. Suomessa käytettävien polttoaineiden ominaisuuksia (Properties of fuels used in Finland). Research Notes of Technical Research Centre of Finland. Espoo: Technical Research Centre of Finland; 2000.
] Schulze WJ. Combustion Control and Burner Management. In: Elliot TC, Chen K, Swanekamp RC, editors. Standard Handbook of Powerplant Engineering. New York: McGraw Hill; 1997. p. 5.33-5.45
0] Jegoroff M, Polttoaineen syöttö ja sen vaikutukset leijukerrospoltossa (Fuel input and its effect on fluidized bed combustion). In: Koskelainen L, Roiha I, editors. Proceedings of Kattilapäivät seminar. Finland: Lappeenranta University of Technology; 2004.
1] Joronen T. Bio- ja seospolttoaineiden syötön ja lämpöarvon hallinta automation avulla. (Control of fuel input and heating value of bio- and mixture fuels by means of automation). In: Koskelainen L, Roiha I, editors. Proceedings of Kattilapäivät seminar. Finland: Lappeenranta University of Technology; 2004.
2] Wimmerstedt R. Recent advances in biofuel drying. Chemical Engineering and Processing 1999; 38: 441-447.
arju K, Kuoren ja puujätteen poltto (Combustion of bark and waste wood). In: itor. Puumassan valmistus osa 2 (Pulp making part 2). Finland: Suomen paperi-
distys; 1983. p. 1519-1582. 4] Wimmerstedt R, Hallsröm A. Torkning av torv och skogsbränslen: teknik, ekonomi,
niques, economy, development needs). Lund
ody-esearch Centre of Finland.
[16] s and economy of the artificial drying of wood chips). SITRA Research Report No. 25 A.
[17]
els : Värmeforsk; 2004.
orkar (Operation .
ale 103-112.
[1] The Finnish Bio Energy Society. Bioenergy terms.
http://www.finbioenergy.fi/default.asp?init=true&InitID=398;0, February 28, 2006. [2] The Finnish Ministry of Trade and Industry. Renewable energy sources and peat.
http://www.ktm.fi/index.phtml?l=en&s=179, February 28, 2006. [3] Alakangas E. Properties and standadization of solid fuels. In: Alakangas E, Agar D, Leppänen M,
editors. Report of the seminar presentations and study tour. Jyväskylä, Finland: OPET CHP/DHC; 2003.
[4] Vanhatalo J, Rautanen J, Ahtila P. Metsäteollisuuden energiantuotannon hiilidioksidipäästöjen vähentäminen biomas
[5
[6
[7
[8
[9
[1
[1
[1
[13] Pellikka J, SavihVirkola NE, edinsinöörien yh
[1utvecklingsbehov (Drying of peat and biofuels: TechUniversity Report, LUTKDH/ TKKA-3002. Lund, Sweden: Lund University; 1984.
[15] Tapanainen J. Turpeen ja puubiomassan kuivaus: kirjallisuuskatsaus (Drying of peat and wobased biomasses: Literature review). Research Notes of Technical REspoo, Finland: Technical Research Centre of Finland; 1982. Linna V, Järvinen T. Hakkeen keinokuivauksen tekniset ratkaisut ja taloudellisuus (Technical solutionHelsinki, Finland: SITRA, The Finnish Ministry of Trade and Industry; 1983. Magnusson L. Uppföljning av erfarenheter från anläggningar för torkning av biobränsle (Follow-up of operation experiences of biofuel dryers). Värmeforskrapport 479. Stockholm, Sweden: Värmeforsk; 1993.
[18] Johansson I, Larsson S, Wennberg O. Torkning av biobränslen med spillvärme (Drying of biofuutilizing waste heat). Värmeforskrapport 881. Stockholm, Sweden
[19] Berge C, Dejfors C. Drifterfarenheter från ångtorkar och direkta rökgastExperiences of Biofuel Dryers). Värmeforskrapport 681. Stockholm, Sweden: Värmeforsk; 2000
[20] Berghel J, Renström R. Basic design criteria and corresponding results performance of a pilot-scfluidized superheated atmospheric condition steam dryer. Biomass&Bioenergy 2003; 23:
57
[21] Hovmand S, Fluidized bed drying. In: Mujumdar AS, editor. Handbook of industrial drying. NeYork: Marcel-Decker;1995. p. 195-248. Wimmerstedt R. Drying o
w
[22] f peat and biofuels. In: Mujumdar AS, editor. Handbook of industrial
[23] ents in Thermochemical Biomass Conversion. Banff, Canada; 20-24 May, 1996.
[25] d pinch technology. In:
[26] r. In: Handbook of Industrial Drying. New York: Marcel-Decker;1995. p. 1-39.
ctor Thesis. Espoo, Finland: Helsinki University of Technology; 2003. ary
[29] m Secondary Heat Flows of the UPM-Kymmene Kaukas
[30] inki University
[31] er.
ust is
[33] apport 671.
] for power l 2004; 83: 773-779.
oenergy 6.
[36]
, eedings of the 10 International Drying Symposium,
[38] n of Drying Properties of Biofuels. In: mass
ss Transfer 1997; Vol. 40 No.12: p 2843-
[40] .
[42]
drying. New York: Marcel-Decker; 1995. p. 809-859. Fagernäs L, Sipilä K. Emissions from Biomass Drying. Proceedings of Conference on Developm
[24] Spets JP, Ahtila P. Reduction of organic emissions by using a multistage drying system for wood-based biomass. Drying Technology: 2004; 22: 541-561. Parviainen K, Fogelholm C-J, Manninen J. Secondary heat systems anGullichsen J, Fogelholm C-J, editors. Chemical Pulping. Helsinki, Finland: Papermaking Science and Technology;1999. p. 336-374. Mujumdar AS, Menon AS. Drying of Solids: Principles, Classification, and Selection of DryeMujumdar AS, editor.
[27] Spets J-P. Enhancement of The Use of Wood Fuels in Heat and Power Production in Integrated Pulp And Paper Mill. Do
[28] Olsson F, Hansson H, Egard M, Bärring M. Tillvaratagande av spillvärme (Utilization of secondheat ). Värmeforskrapport 824. Stockholm, Sweden: Värmeforsk; 2003. Ahtila P, Svinhufvud P. Heat Recovery froMill. European Conference on Industrial Energy, Vienna; 1998. Kilponen L, Ahtila P, Parpala J, Pihko M. Improvement of pulp mill energy efficiency in an integrated pulp and paper mill. Research report TKK-EVO-A5. Espoo, Finland: Helsof Technology; 2000. Renström R, Berghel J. Drying of Sawdust in an Atmospheric Pressure Spouted Bed Steam DryDrying Technology 2002; 20: 449-464.
[32] Berghel J, Renström R. Controllability of Product Moisture Content When Non-screened SawdDried in a Spouted Bed. Drying Technology 2004; 22: 507-519. Axby F. Styrning av biobränsletorkar (Control of biomass dryers), VärmeforskrStockholm, Sweden: Värmeforsk; 1999.
[34 Nyström J, Dahlqvist E. Methods for determination of moisture content in woodchips plant – a review. Fue
[35] Impola R, Saastamoinen J, Fagernäs L. Kiinteiden polttoaineiden kuivumisparametrien karakterisointi (Characterization of drying parameters for solid fuels). Final Report of BiResearch Programme-301. Jyväskylä, Finland: Technical Research Centre of Finland; 199Saastamoinen J, Impola R. Drying of Solid fuel Particles in Hot Gases. In: Rudolph V, Keey RB, Mujumdar AS, editors. Proceedings of the 9th International Drying Symposium, Gold Coast, Australia; 1-4 August, 1994. p. 623-630.
[37] Saastamoinen J, Impola R. Drying of Biomass Particles in Fixed and Moving Beds. In: Strumillo CPakowski Z, Mujumdar AS, editors. Proc th
Krakow, Poland; 30 July-2 August, 1996. p. 349-356. Impola R, Saastamoinen J, Kääriäinen J. CharacterizatioChartier PH, Beenackers AACM, Grassi G, editors. Proceedings of the 8th European BioConference, Vienna, Austria; 3-5 October, 1994. p. 1498-1505.
[39] Johasson A, Fyhr C, Rasmusson A. High temperature convective drying of wood chips with air andsuperheated steam. International Journal of. Heat and Ma2858. Perre P, Morse M, Martin M. Advances in transport phenomena during convective drying superheated steam and moist air. International Journal of Heat and Mass Transfer 1993; Vol. 36: No11: p. 2725-2746.
[41] Wang ZH, Chen G. Heat and mass transfer in fixed-bed drying. Chemical Engineering Science 1994;54: p. 4233-4343. Keey RB. Drying: Principles and Practice. International series of monographs in chemical engineering, 13. Oxford; 1972.
[43] Lydersen AL. Mass Transfer in Engineering Practice. Chichester, New York, Brisbane, Toronto, Singapore: John Wiley&Sons; 1983.
58
[44] Lampinen MJ. Kemiallinen termodynamiikka energiatekniikassa (Chemical ThermodynaEnergy Engineering). Publications of Laboratory of Applied Thermodynamics, Espoo, Finland:Helsinki University of T
mics in
echnology; 1996.
ja palaminen. Jyväskylä, Finland: International Flame
[46] e 1 Nordic Drying Conference,
[47] neering. Boston, London, Singapore, Sydney, Toronto,
[48] ited Kingdom: Institution of Chemical Engineers; 1998.
[45] Hyppänen T, Raiko R. Leijupoltto (Combustion in fluidized bed). In: Raiko R, Kurki-Suonio I,Saastamoinen J, Hupa M, editors. PolttoResearch Foundation, Suomen kansallinen osasto; 1995. Spets J-P, A new multi-stage drying system. In: Proceeding of th st
Paper No. 13. Trondheim, Norway; June 27-29, 2001. Kunii D, Levenspiel O. Fluidization EngiWellington: Buttenworth-Heinemann; 1991. Brennan D. Process Industry Economics. Un
[49] Nord Pool. Elspot monthly prices. http://www.nordpool.com/nordpool/spot/index.html, F2006. The Finnish Ministry of Trade and Industry. Energiakatsaus (Energy Review). Helsinki, F
ebruary 28,
[50] inland:
[51] 19(8):
[53] atuohje (Standard Instruction for Bio Fuel).
[54] editor. . p. 71-111.
The Finnish Ministry of Trade and Industry, Energy Department; 4/2005. Menshutina NV, Kudra T. Computer Aided Drying Technologies. Drying Technology 2001; 1825-1849.
[52] Jumah RY, Mujumdar AS. Control of Industrial Dryer. In: Mujumdar AS, editor. Handbook of Industrial Drying. New York: Marcel-Dekker; 1995. p. 1343-1367. The Finnish Bio Energy Society. Puupolttoaineiden laPublications of Finbio. Finland: The Finnish Bio Energy Society; 1998. Pakowski Z, Mujumdar AS. Basic Process Calculations in Drying. In: Mujumdar AS, Handbook of Industrial Drying. New York: Marcel Dekker; 1995
[55] Kotas TJ. The Exergy Method of Thermal Plant Analysis. Malabar, Florida: Krieger PublishingCompany; 1995.
ISBN 978-951-22-8648-5ISBN 978-951-22-8649-2 (PDF)ISSN 1795-2239ISSN 1795-4584 (PDF)